Formal mentoring versus supervisor and coworker relationships ...

birdsloafInternet and Web Development

Jul 5, 2012 (4 years and 9 months ago)


Journal of Organizational Behavior
J.Organiz.Behav.24,271–293 (2003)
Published online in Wiley InterScience (
Formal mentoring versus supervisor
and coworker relationships:differences
in perceptions and impact
Siemens Qualification and Training,Munich,Germany
Central Michigan University,Mount Pleasant,Michigan,U.S.A.
Summary Formal mentoring programs in two companies were examined regarding (1) the extent to
which mentees and mentors agreed on the nature of the mentoring relationships and (2) the
extent to which dimensions of mentoring relationships were related to outcomes for the men-
tees,compared with the extent to which dimensions of supervisory and coworker relationships
were related to the same outcomes:job satisfaction,organizational commitment,and turnover
intentions.Mentors were at least two hierarchical levels above the mentee,and both were part
of the companyies’ formal mentoring program.Sixty-one pairs of mentors and mentees
participated.Overall,there was little agreement between mentees and mentors regarding
the nature of the mentoring relationship.Furthermore,the mentoring relationship was not
related to mentee outcomes,while supervisory and coworker relationships were.It is
suggested that,if one desires to affect job satisfaction,turnover intentions,and organizational
commitment,mentoring functions may be best performed by supervisors and coworkers
rather than assigned formal mentors fromhigher up in the organizational hierarchy.Copyright
#2003 John Wiley & Sons,Ltd.
Mentoring is traditionally defined as developmental assistance offered to a junior employee by some-
one more senior and experienced in the organization (e.g.,Kram,1983),and the use of formal mentor-
ing programs is likely to increase (Tyler,1998) because of their presumed positive impact on the
mentee.In order to provide the benefits that mentoring relationships bring,many companies already
have established formal mentoring programs.The belief that informal mentoring is better,deeper or
more valuable than formal mentoring is still a viable alternative,although the few studies that address
this issue leave room for doubt (e.g.,Chao,Walz,& Gardner,1992;Ragins,Cotton,& Miller,2001).
Formal mentoring relationships could be more superficial because formal mentoring attempts to
legislate interpersonal chemistry and personal commitment (Kram,1985).Therefore,in a mandated
formal mentoring relationship,the mentor’s motivation and the mentee’s openness might be lower
Received 4 February 2002
Revised 23 September 2002
Copyright#2003 John Wiley & Sons,Ltd.Accepted 15 January 2003
*Correspondence to:Babette Raabe,Siemens Qualification and Training,St.-Martin-Str.76,81541 Munich,Germany.
(Chao,Walz,& Gardner,1992).In spite of their popularity,however,little is known about the effects
of formal mentoring programs (Ragins,Cotton,&Miller,2000).As Scandura (1998) puts it:‘To sum-
marize,the jury is still out on the efficacy of formal mentoring programs.’ Since the mid-1990s the
traditional concept of mentoring has actually broadened,and interest in other forms of mentoring rela-
tionships such as peer and team mentoring (see Eby,1997) increased.
Kram’s (1983) work was influential in directing early mentoring research and practice.She defined
mentoring as role activities that senior,experienced organizational members assume with junior
members,and she posited that the mentoring relationship has two major dimensions:work-related
developmental functions and psychosocial support functions.The parallel with a traditional two-
dimensional taxonomy of leadership (work-related and person-related;Blake & Mouton,1982;
Fleishman & Harris,1962;Yukl & Van Fleet,1992) is striking.Scandura and Ragins (1993) added
a third mentoring dimension:role modeling.
Thus,mentoring is often described as consisting of three behaviors making up relationships between
mentors and mentees:career development,social support,and role modeling.Research suggests that
mentors help mentees in at least two domains—social (e.g.,Thomas,1993) and career (e.g.,Fagenson,
1989;Kram,1985;Riley &Wrench,1985)—and thus should make the employees more valuable to the
organization as well as helping the employees’ careers.Three organizational outcomes affected by
mentoring are job satisfaction,organizational commitment,and retention or reduced turnover (Aryee,
Chay,& Chew,1996;Chao et al.,1992;Corzine,Buntzman,& Busch,1994;Fagenson,1989;Goh,
1991;Mobley,Jaret,Marsh,&Lim,1994;Ragins &Cotton,1999;Seibert,1999;Whitely &Coetsier,
1993),and the present study focused on these outcomes.The present study addressed two questions
regarding formal mentoring programs.Firstly,because the mentee’s own perceptions of mentoring rela-
tionships are likely to impact the mentee,we wondered how closely the perceptions of the two most
involved parties,the mentor and the mentee,are aligned.Many authors have argued for the importance
of mentors’ and mentees’ perceptions (see reviewand model by Young &Perrewe
,2000a).Would men-
tors and mentees perceive certain behaviors in the relationship in the same way?Some researchers have
pointed out that a certain degree of agreement in leadership perceptions of mentors and proteges
(Atwater & Yammarino,1992;Wohlers & London,1989) is an important factor for the quality of
the mentoring relationship,while others found that perceptual agreement was not necessarily related
to higher effectiveness ratings of the mentoring relationship (Godshalk & Sosik,2000).
Mentoring interactions are based on chains of reactions and counter-reactions (Jones & Gerard,
1967) as well as thoughts,feelings,intentions,and plans of each of the participants (Harre
& Secord,
1972).Meeting the other’s expectations plays an important role (Young &Perrewe
standings between mentor and mentee therefore should lower the quality of the mentoring relationship,
and therefore the first research question was as follows:
I.To what degree do mentors and mentees agree in their perceptions about the mentoring behaviors
or the nature of the mentoring relationship?
Secondly,in answering the question about the effectiveness of mentoring relationships,we sought
comparisons.That is,howdoes the mentoring relationship compare to other potentially important rela-
tionships in the workplace with regard to certain outcomes?In principle,activities and functions with
outcomes similar to mentoring might be performed by people other than one’s formal mentor.These
are conceived as mentoring substitutes:
II.How well do the relationships with the various potential mentors (formal mentors,supervisors,
and coworkers) relate to mentee outcomes (organizational commitment,job satisfaction,and turn-
over intentions)?
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
Regarding the first question,most mentoring research relied upon mentee’s descriptions (i.e.,per-
ceptions) of the mentor’s behavior and the mentor–mentee relationship (Higgins &Kram,2001).This
makes sense when looking at outcomes that are under the mentee’s control (i.e.,mentee reactions such
as satisfaction,commitment,and turnover),because the mentee’s own perceptions and beliefs about
the mentoring relationship should most directly affect his/her own behavior.Because previous research
relied on a single perspective to describe the mentoring relationship (Higgins &Kram,2001),however,
we do not know the extent to which the mentor and mentee have similar perceptions of their relation-
ship (see Godshalk & Sosik,2000,for an exception).Much of the theoretical base for the mentoring
relationship relies on social exchange theory (Blau,1964;Homans,1958;Thibaut & Kelley,1959).
Coming froma social exchange perspective,mentors and mentees exchange certain things within their
reciprocal mentoring relationship,and potential benefits for both mentees and mentors are often noted
(Ragins & Scandura,1999;Mullen,1994;Kram,1985;Phillips-Jones,1982).But is the mentoring
relationship primarily one of the perceptions and feelings the mentee has about it,or is it truly a
mutually shared,interactive experience?Do both mentor and mentee perceive the situation in the same
way?The present study allowed the examination of these issues by measuring the mentoring relation-
ship from both the mentees’ and mentors’ perspectives.
The criterion variables chosen for the addressing the second question were turnover intentions,
organizational commitment,and job satisfaction,because they fit with the social exchange theme.
Currently there are multiple versions of social exchange theory (Cropanzano,Rupp,Mohler,
& Schminke,2001).Originally,it was consistent with the specific idea that social relationships were
akin to marketplace exchanges.Thus,one was likely to continue in the relationship if one’s exchange
with the ‘other’ was favorable;i.e.,if one received more fromthe other person than one gave,there was
sufficient reason to stay in the relationship (e.g.,Thibaut &Kelley,1959).Although different versions
of the theory evolved so that it is sometimes more generally about satisfying qualities of relationships,
the basic idea is still that favorable social relationships provide an incentive to stay in these relation-
ships.It is important to note that the mentors in the present study were likely to represent the organiza-
tion in the mentees’ perceptions for two reasons.First,they were relatively high in the organization’s
hierarchy (and at least two levels above the mentees),and second,the organization sponsored the men-
toring as an official,formal program.It was reasonable,therefore,for the mentees to consider the men-
toring exchange to be between themselves and their organization,as well as with the specific people
who were their mentors.Given this,intentions to remain in the relationship with the organization and
to continue exchanging with it (turnover intentions and organizational commitment) were particularly
salient criteria.Job satisfaction was included as the third criterion,because the favorable mentoring
relationships should result in favorable attitudes toward the workplace.
The second question concerns the identity of the mentor and is related to the issue of formal and
informal mentoring.In the present study all participants had formal mentors.There are many activities
and many sources of employee development (Maurer,Pierce,& Shore,2002).In addition to formal
mentoring,other people typically have important developmental relationships with the employee.
Who is the mentor?Kram (1983) defined mentoring as role activities in which senior,experienced
organizational members engage.In more recent years,however,the mentoring concept moved away
from the typical mentoring relationship between an older,more experienced and a younger,less
experienced person to different concepts and kinds of mentoring.Besides a mentee’s classic relation-
ship with a mentor a couple of steps above him/her in the organizational hierarchy,mentoring by the
direct supervisor is often considered.In one recent study (Ragins,Cotton,&Miller,2000),a little over
50 per cent of the mentors were direct supervisors.In addition,more recent suggestions for sources of
mentoring include peers,groups,and even subordinates (Russell &Adams,1997).Classically,the for-
mal mentors are one or more levels above the supervisor in the same organization and are more senior
(Kram,1983).In addition,a good mentor should have certain personal characteristics (e.g.,know the
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
company,have some influence,be interpersonally competent).The formal mentoring programs in the
current study tried to use these selection criteria for formal mentors.
More recently,it was argued that mentoring can be done by different people or even multiple people
(e.g.,Higgins & Kram,2001;Higgins & Thomas 2001;Russel & Adams,1997;Eby,1997).Even
when there is a formal mentor,it is still possible that others in the organization can performsome psy-
chosocial or career enhancement mentoring functions as well.Higgins and Kram (2001,p.265)
remarked that ‘individuals beyond a primary senior person seldomhave been considered’ (as mentors)
in research on mentoring.This suggests that separate sources of developmental workplace relation-
ships need to be examined in a single study so that they can be compared.
Research conceiving supervisors and coworkers as potential mentors encounters a measurement
dilemma.The leadership and mentoring literatures,for example,developed separately from each
other.As a result,the typical supervisor (leadership) dimensions found in the research on leadership
are not identical to the mentoring dimensions typically found in the mentoring literature,even though
there are definite similarities.We decided to examine variables that come fromthe separate leadership
and mentoring literatures.Leader–Member–Exchange (LMX) theory is one of the most clearly dyadic
leadership theories,and mentoring is also a dyadic function.Therefore,we measured LMX dimen-
sions as the supervisor dimensions,and compared their relationship to employee outcomes with the
mentor’s relations to employee outcomes.
Like the origins of mentorship theory,LMX theory is based on the concept of social exchange,pro-
viding a notable parallel between mentoring and leadership processes.The dyadic nature of the LMX
relationship can vary from one subordinate to the next for the same supervisor (Danserau,Graen,&
Haga,1975;Yukl &Van Fleet,1992).This leadership theory (LMX) is applied primarily to situations
in which the direct supervisor is the leader of one or more subordinates (who report directly to this
supervisor).Supervisors,different people from mentors in the present study,have responsibility for
the progress and work of the employee as well as the conformity of the employee’s behavior to com-
pany policy.Supervisors are often considered part of management,and that was true in the present
In the leadership literature,supervisors are often examined for their influence on some of the same
subordinate outcomes that mentors might also influence,such as providing supportive behaviors at the
workplace (Higgins & Thomas,2001).Indeed,many of the mentoring functions also resemble the
essence of what is labeled social support in the occupational stress literature.Supervisors,as well
as formal mentors,can influence outcomes such as job satisfaction and organizational commitment
(Gerstner & Day,1997).LMX theory of leadership focuses directly on one-to-one relationships
between supervisors and subordinates,much like mentoring.This encouraged our decision to focus
on supervisor behaviors found in LMXtheory for comparisons of supervisor relationships with formal
mentor relationships.
McManus and Russell (1997) suggested two important similarities between LMX and mentoring:
Both are developmental relationships in the workplace,and both are subject to a role-making
process connected to a negotiating latitude about the nature of the relationship.Indeed,sometimes
supervisors are even formally considered to be mentors and perform mentoring functions (Scandura
& Schriesheim,1994;Tepper,1995).In order to examine differences between the supervisory and
mentor dyadic relationships in the present study,a clear-cut definition of the mentor was applied that
prevented the mentoring data from being confounded with supervisor developmental influences.
That is,we studied mentoring in organizations that had formal mentors who were not the mentees’
direct supervisors.In this way,supervisors’ potential effects on subordinates’ outcomes could be
examined separately from mentors’ effects.
As noted above,mentoring and LMX both are based on social exchange theory,and each is com-
posed of both work-related and person-related interactions.The mentoring dimensions in the present
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
study are psychosocial support,career development,and role modeling,while the LMXdimensions of
leadership or supervision are contribution,affect,loyalty,and professional respect.Two of the men-
toring dimensions are quite similar to two of the supervision dimensions,while one mentoring dimen-
sion and two supervisory dimensions do not have as direct parallels in the other domain.The
mentoring psychosocial support dimension is an expression of positive,friendly affect,which finds
a parallel in the LMXsupervisory affect dimension.An operational difference is that some of the men-
toring items are more specific activities (e.g.,‘I often go to lunch with my mentor’),while the LMX
items are more general (‘my supervisor is a lot of fun to work with’).The mentoring role modeling
dimension (e.g.,‘I respect my mentor’s knowledge of the profession’) has a parallel in the LMX pro-
fessional respect dimension (e.g.,‘I admire my supervisor’s professional skills’),because both are
expressions of respect and admiration.In addition,LMXcontains contribution and loyalty dimensions
and mentoring includes the career development dimension.Of these,both the mentors’ career devel-
opment dimension and the supervisor (LMX) loyalty dimension concern helping the employee in his
or her career in the organization,but operationally the items suggest doing this in very different ways.
Mentoring career development appears to represent positive proactive behavior (e.g.,‘my mentor
advised me about promotional opportunities’),while supervisor (LMX) loyalty is more protective
and defensive of the subordinate (e.g.,‘my supervisor would defend me to others in the organization
if I made an honest mistake’).Thus,there is clearly some overlap in the constructs and measures of
mentoring and LMX relationships,but there are also some differences.
The conceptual and operational similarities suggest that the dimensions of the mentoring and super-
visor measures should affect some of the same outcomes.The differences in relationships of similar
mentoring and supervisory dimensions with outcomes should be due mostly to the source of the rela-
tionship (mentor versus supervisor).This would be especially true for the most similar dimensions
(e.g.,mentor role modeling and supervisor professional respect should be similarly related to out-
comes) and least true where there is least similarity (e.g.,supervisor loyalty and any mentoring dimen-
sion would not be similarly related to outcomes).All items in each dimension of the mentoring and
LMX subscales can be seen in the Appendix.
Psychosocial support and career development by the mentor should be experienced as beneficial to
the mentees and should make them more satisfied,more committed,and less likely to turn over (as
proposed by Young &Perrewe
,2000a).Support is positive interpersonal contact and should be experi-
enced as pleasurable.Career development by the mentor gives the mentee a special advantage in the
organization and should also lead to more commitment.Role modeling represents respect for the men-
tor and may also result in satisfaction and reluctance to depart fromthe organization.Indeed,as noted
earlier,previous research has often found mentoring to be related to satisfaction,commitment,and less
turnover (Aryee et al.,1996;Chao et al.,1992;Corzine et al.,1994;Fagenson,1989;Goh,1991;
Mobley et al.,1994;Ragins & Cotton,1999;Seibert,1999;Whitely & Coetsier,1993).Regarding
supervision,affect,loyalty and professional respect bear some resemblance to the mentor dimensions,
and the same rationale should apply.They are likely to be pleasurable relationships and should also
result in satisfaction,commitment,and retention.Contribution,the other dimension in LMX,is an
expression of willingness to expend extra effort for the supervisor,an indicator of positive regard
for the work situation.There is a long history of research on leadership or supervision,and it usually
concludes that supervisors can influence attitudes and actions of subordinates (Yukl & Van Fleet,
1992).Regarding LMX in particular,the research history is for a shorter period,but it also indicates
the same type of influence (Graen & Uhl-Bien,1998).
Mentors’ primary goals are supposed to be developing the mentees for career growth (e.g.,Kram,
1983).Supervisors are expected to get more immediate results from their subordinates,but employee
development is also one of their goals (e.g.,McManus &Russell,1997).Aside fromgoals,similarities
in certain aspects of the relationships themselves are apparent.Both relationships are characterized by
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
a work focus and a person focus,e.g.,initiating structure and consideration by supervisors (Fleishman
& Harris,1962),and help in the social and career domains by the mentor (e.g.,Kram,1985;Thomas,
Coworkers,by comparison,appear to have quite different goals and relationships with each other,
although a study by Maurer and Palmer (1999) is suggestive that exchange might also be a relevant
topic when we consider employee development in relation to coworkers.Coworkers are hierarchical
peers with the focal person,unlike most formal mentors and supervisors.They also would not usually
see it as part of the their official role to direct the work of and develop the skills and careers of their
coworkers.Any such effects they have would therefore be secondary and perhaps even accidental.
Nevertheless,their effects on colleagues or peers can be strong and are effected through multiple
means (e.g.,through both discretionary and ambient stimuli;Hackman,1992).Peers can,for example,
directly offer advice and information on how to accomplish goals,inform each other of potential
chances for advancement,and socially reinforce either good or bad work behaviors.The influence
of all three on the reactions of the focal person can of course be informal;people are social beings
after all.In addition,however,the formal mentor who is at least two positions above the focal person
in the organization and the supervisor have the added influence of hierarchical advantage.The cow-
orker does not have any formal advantage over the person and is likely to wield only informal social
influence.In this type of work situation,however,the formal mentor is likely to have one disadvantage
compared to supervisors and coworkers in wielding influence:He or she is likely to have less frequent
contact with the person.Finally,comparing the three positions in relation to a focal person,the super-
visor might have one advantage over the other two.The supervisor is more likely than the others to
have routine formal input into the employee’s rewards,because he or she is likely to be involved in
performance appraisal and reward systems regarding the employee.These similarities and differences
among mentor,supervisor,and coworker were true in the companies in the present study,but they are
also likely to be true in most companies.
This study patterned the approach to coworker relationships after the supervisor relationships.
Although there is not as much research on coworker influences (as there is on leaders or supervisors)
historically,it is well accepted that coworkers also influence employees’ reactions to the workplace
(e.g.,Riordan & Griffeth,1995;Major,Kozlowski,Chao,& Gardner,1995;Revicki,Whitley,&
Gallery,1993).As Kramstated,‘peer relationships appear to offer a valuable alternative to the mentor
relationship;they can provide some career and psychosocial functions,they offer the opportunity for
greater mutuality and sense of equality,and they are more available in numbers.’ (Kram,1983,p.623).
The concept of mentoring as psychosocial support and developmental assistance applies to coworkers
as well as to supervisors.When examining possible coworker influence on outcomes,the choice of
dimensions to measure was less certain than with supervisors.In the past,much peer research focused
on entire groups (e.g.,group cohesiveness,Podsakoff,MacKenzie,& Ahearne,1997;Zacarro &
Dobbins,1989;Griffeth,Hom,& Gaertner,2000;group climate,Anderson & West,1998;Zohar,
2000) rather than on a dyadic relationship as mentoring research does,and therefore it does not always
provide a good comparison with mentoring measures.Thus,similar to a recent study by Sherony and
Green (2002),we decided to adapt the measures from the LMX dimensions to measure the relation-
ships of the focal employee with his or her single most-liked coworker.
Overall,therefore,this study’s purposes were (1) to determine the strength of the relationship
between the formal mentor’s and mentee’s perceptions of their relationship with each other and (2)
to compare mentor–mentee relationships with supervisor–subordinate relationships and with cowor-
ker–coworker relationships in relation to mentee outcomes.Finding strong relationships between per-
ceptions would be favorable,because movement toward the formal mentoring program’s goals should
be facilitated if both parties know the current situation.Without agreement,however,it would be
less likely that the mentor and mentee would see the same steps needed to achieve the goals.Formal
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
mentoring programs,by bringing the two parties together and training them,should facilitate shared
agreement on where the relationship is and where it needs to go.Furthermore,by examining formal
mentors,and supervisors and coworkers who were not formal mentors in the same sample,the present
study was able to compare their relative effects for the first time.It is important that formal mentoring
programs provide added value to the development of the mentee.This means that they should have an
effect on mentees that is additional to any effects by the supervisor and coworkers.The present
study,therefore,examined and compared relationships of formal mentors,supervisors,and coworkers
with regard to three employee outcomes:job satisfaction,organizational commitment,and turnover
Sample and procedure
Employees in two formal mentoring programs that were developed and overseen by outside consul-
tants were the focal people or mentees in the study.The data for the present study were part of a larger
dataset on relationships in the organizations.One company was in the energy industry in the south-
eastern United States,and the other was a high-technology company in the northwestern United States.
The mentoring programs identified and trained as mentors people who were at least one hierarchical
level above the mentees’ supervisors in the company.
The mentoring programin the northwestern technology company was 2 years old and originally had
a diversity goal to retain females and minorities.At the time the study took place,however,the pro-
gram had been expanded recently to cater to all employees.Program goals for the organization
included retention of valued employees,increased productivity,and acceleration in employee
development;for employees,goals included support to achieve career self-reliance,enhancing career
development,and individual career growth.Participation in the programwas voluntary and followed a
A consulting firm was hired to set up,train,and implement a mentoring system in these two com-
panies.We were third parties,with no prior links either to the companies or the consulting firm,but
they were open to letting us observe and obtain the data for this research.Although we carried out
the study at our own expense,both the consulting firm and the companies were helpful and coop-
erative.It was 1997,and both companies were healthy,taking part in the U.S.economic boomof the
1990s.They were both interested,however,in improving the career prospects of their female and
minority professionals and managers,and this was part of their motivation for establishing the men-
toring program.The programwas not just beginning,but was already established at the time of the
study (about 1 year old in one company and 2 years old in the other).Our liaison managers in the
firms were promised that the firms would receive a report about the general quality of relationships
of mentees with their mentors,supervisors,and coworkers,and they received that report in both
written and oral form.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
process involving applications,the fulfillment of certain eligibility requirements (e.g.,minimumorga-
nizational tenure and stated commitment to participate in mentoring programactivities),and a match-
ing of goals of the mentees and skills and background.Mentees and mentors took part in an orientation
meeting,received separate training,and had to commit to a minimum time investment of 2 hours per
month and to program evaluations;mentoring program managers gave support,if necessary.Asked
what helps most in order to establish a good mentoring relationship,mentoring program managers
answered that mentees need to be committed to the mentoring relationship and their own specific
development goals,to act goal- and results-driven,and to take initiative.Further comments were that
both sides need to build trust and open communication,and that a good match between mentor and
mentee is important for the relationship success.Challenges that the programs faced referred to fund-
ing sources in the organization and upper management support as well as time constraints of both men-
tees and mentors that had to be overcome.
The mentoring programin the southeastern energy company was somewhat younger,also had diver-
sity roots in terms of retaining African–Americans and females,and had been implemented 6 months
prior to the study.The duration of the programwas 12 months.Organizational goals were to integrate
minorities and women into the company culture,as well as retention,commitment,and improved
recruiting efforts in the long run.For mentees,mentoring program goals were professional growth
and development through the attainment of personal and professional goals.The program followed
the same routine as in the other company:orientation,training,and ongoing support in the mentoring
relationship by the mentoring program managers,and program evaluations.The mentoring program
manager stated that willingness of the participants in terms of a strong commitment as well as an ela-
borate matching process helped most in order to establish a good mentoring relationship.This pro-
gram’s challenges also pertained to inconsistent upper management support,where much political
energy was spent to ensure the existence of the program.
One hundred and seventy-five mentees in the programs were asked to participate in the study,and 87
did so,for a response rate of 49 per cent.All of their mentors were asked to participate by completing a
questionnaire,and 109 did so,for a response rate of 57 per cent.The present study analyzed data only
frommatched pairs,which totaled 61 pairs for the largest analysis with no missing data.Mentees were
given the questionnaires in person or via company mail,while all of the mentors received theirs
through company mail.All participants returned their questionnaires to the consultants through com-
pany mail.The mentees’ average age was 34.7 years,61 per cent were women,77 per cent had bache-
lors or graduate degrees,62 per cent reported that they were Caucasian,and they had been in their
current jobs for an average of about 34 months (and in the company for 85 months).The mentors’
average age was 44.1 years,40 per cent were women,96 per cent had bachelors or graduate degrees,
86 per cent reported that they were Caucasian,and they had been in their current jobs for an average of
about 47 months (and in the company for 203 months).
Unless otherwise indicated,all items of the measures were answered on seven-point,agree–disagre
scales.Alphas were derived from the sample of the 61 mentee–mentor dyads.Items in the mentoring
and supervisory scales are in the Appendix.The items for the coworker scales are parallel to the super-
visory items.
Mentee questionnaires
The mentees’ questionnaires measured their relationships with their mentors,their supervisors,and
their best-liked coworker,as well as the outcome variables job satisfaction,organizational commit-
ment,and turnover intent.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
Three mentoring relationships were measured:Career development (6 items,M¼4.98,SD¼
0.94,alpha ¼0.73) provided by the mentor,psychosocial support activities (5 items,M¼4.28,
SD¼1.07,alpha ¼0.65) of the mentee toward the mentor,and role modeling (4 items,M¼5.62,
SD¼0.86,alpha ¼0.78) provided by the mentor.These measures were developed by Scandura and
Ragins (1993).Asample itemfromthe career development subscale was ‘My mentor has placed me in
important assignments,’ a sample itemfromthe psychosocial support subscale was ‘I socialize with my
mentor after work,’ and a sample itemfromthe role modeling subscale was ‘I try to model my behavior
after my mentor.’
Mentees rated their relationships with their supervisors on four LMX dimensions:Affect for the
supervisor (3 items,M¼5.64,SD¼0.97,alpha ¼0.84),professional respect for the supervisor
(3 items,M¼5.78,SD¼1.26,alpha ¼0.94),loyalty of the supervisor to the mentee (3 items,
M¼5.45,SD¼1.12,alpha ¼0.83),and contribution of the mentee to the supervisor (4 items,
M¼6.16,SD¼0.57,alpha ¼0.61).These measures were from the LMX scale of Liden and Maslyn
(1998).A sample item for affect for supervisor was ‘I like my supervisor very much as a person,’ a
sample item for respect was ‘I respect my supervisor’s knowledge of and competence on the job,’
a sample itemfor loyalty was ‘My supervisor would defend me to others in the organization if I made
an honest mistake,’ and a sample item for contribution was ‘My supervisor can depend on me when
we are overloaded with work.’
Mentees rated their relationships with their coworkers with items written to be parallel with the
supervisor items,and therefore parallel indices were developed:Affect for the coworker (M¼6.19,
SD¼0.68,alpha ¼0.76),professional respect for the coworker (M¼6.32,SD¼0.62,alpha ¼0.86),
loyalty of the coworker to the mentee (M¼5.70,SD¼0.96,alpha ¼0.82),and contribution of the
mentee to the coworker (M¼6.17,SD¼0.56,alpha ¼0.73).The items for these indices were written
to be parallel with the items regarding the supervisor (fromLMX) and can therefore be consider mea-
sures of Coworker–Member Exchange (CMX).
Finally,mentees also answered questions about the study’s outcome variables:job satisfaction,orga-
nizational commitment,and turnover intentions.Job satisfaction (M¼5.42,SD¼1.11,alpha ¼0.87)
was measured by the mean of seven items partly obtained from the Michigan Organizational Assess-
ment Scale (Cammann,Fichman,Jenkins,&Klesh,1979).A sample itemfor job satisfaction was ‘In
general,I like working in this post/unit.’ Organizational commitment was measured by having the
mentees rate 10 items,a sample item of which is:‘I am willing to put in a great deal of effort beyond
that normally expected in order to help this organization be successful.’ The items to assess organiza-
tional commitment were taken from the short form of the Porter Scale (Mowday,Steers,& Porter,
1979) and one item from Nadig (1994).The reliability for this scale was good (M¼5.96,
SD¼0.6975,alpha ¼0.85).
Mentees rated their turnover intent (M¼3.12,SD¼1.37,alpha ¼0.62) with a three-item measure
fromthe Michigan Organizational Assessment Questionnaire (Seashore,Lawler,Mirvis,&Cammann,
1982).A sample item was ‘I plan to look for a new job within the next 12 months.’
Mentor questionnaires
The mentors’ questionnaire measured the same three relationships between mentor and mentee that the
mentee’s questionnaire did,using items that were directly parallel to those in the mentee questionnaire:
career development that the mentor does for the mentee,psychosocial support of the mentor toward
the mentee,and role modeling (the extent to which the mentee follows the mentor’s model).Descrip-
tive data for the subscales are as follows:career development (M¼5.29,SD¼0.67,alpha ¼0.60),
psychosocial support (M¼3.86,SD¼1.05,alpha ¼0.69),and role modeling (M¼5.01,SD¼0.81,
alpha ¼0.83).A sample itemfromthe career development subscale was ‘I have placed my prote
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
important assignments,’ a sample item from the psychosocial support subscale was ‘I socialize with
my protege after work,’ and a sample item from the role modeling subscale was ‘My prote
tries to
model my behavior.’ All the mentoring items,phrased from the mentee’s point of view,are in the
When interpreting the results one has to bear in mind that some of the reliabilities were rather low
(i.e.,in the 0.60s).For the mentee survey,this holds especially true for perceptions of the mentor’s
psychosocial support and contribution to the supervisor,and mentees’ turnover intent.For the mentor
survey,perceptions of career development and psychosocial support obtained low alphas.Thus,some
of the results should be interpreted cautiously,because although they might reflect real phenomena,
they could also be due to measurement error.
Relations between mentor and mentee perceptions
of the mentoring relationship
Because the study was conducted in two different organizations,it was possible that something in the
different contexts in these two organizations might make mentoring,supervision,and coworker rela-
tionships have different effects.The northwestern company came froma young and progressive indus-
try and had a very flexible and innovation-based culture with strong company values and a high
employee orientation and teamwork focus.The southeastern company came from a more traditional
mature industry,had a respectable history of its own,was a key player in its field,and had a caring
approach for its employees.It was very performance-oriented,placed a strong focus on customer satis-
faction,and achieved very good results in that.
Due to the differing company cultures,the relationships with mentors,supervisors,and coworkers
were examined via partial correlations with company as the covariate.In addition,duration of mentor-
ing relationship was also controlled,because such relationships might change over time.All the con-
clusions from these data remained the same regardless of whether these variables were controlled or
not;all the relationships with significant correlations also had significant partial correlations,and all
the relationships with nonsignificant correlations also had nonsignificant partial correlations.This
argues against context having a strong effect on the results.
The mentors and mentees showed no agreement on the scale level regarding the nature of their rela-
tionship when company and length of mentoring relationship were controlled (correlations were 0.21,
0.01,and 0.15 for psychosocial support,career development,and role modeling,respectively;all were
n.s.).On an item level,three partial correlations in the psychosocial support scale,one in the career
development scale,and none in the role modeling scale were significant (see Table 1):sharing personal
problems,socializing after work,considering the other one to be a friend,and the mentor placing
the mentee in important assignments.Therefore,this exploratory analysis indicates that the psychoso-
cial support function might be the one in which there is the best chance for agreement,even though
the results at the scale level were not significant.This also could be concluded by the correlations at
the scale level,where the strongest of the nonsignificant correlations was the one for psychosocial
Comparison of the means (see Table 2) with t-tests suggests that the mentors believed they were
giving a little more career development than mentees believed they were getting,but mentees believed
there was more mutual psychosocial support and that they were modeling their behaviors after the
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
Table 1.Mentor and mentee shared perceptions of the mentoring relationship:partial correlations of mentoring
behaviors (n) at the item level
Psychosocial support (scale)
Sharing personal problems 0.41** (60)
Socializing after work 0.29* (60)
Exchanging confidences 0.22 (60)
Considering the other one to be a friend 0.48*** (60)
Often going to lunch with each other 0.07 (61)
Career development (scale)
Mentor takes personal interest in mentee’s career 0.02 (61)
Mentor placed mentee in important assignments 0.37** (61)
Mentor gives mentee special coaching on the job 0.06 (61)
Mentor advises mentee about promotional opportunities 0.05 (61)
Mentor helps to coordinate professional goals 0.05 (61)
Mentor devoted special time and consideration to mentee’s career 0.14 (61)
Role modeling (scale)
Mentee tries to model behavior after mentor 0.03 (60)
Mentee admires mentor’s ability to motivate others 0.24 (59)
Mentee respects mentor’s knowledge of the profession 0.07 (60)
Mentee respects mentor’s ability to teach others 0.21 (59)
*p <0.05;**p <0.01;***p <0.001.
Note:The variables that were used as covariates are company as well as length of mentoring relationship.
Table 2.t-Tests on mentor and mentee perception differences of mentoring behaviors (SD)
Mentee Mentor t
mean mean
Psychosocial support (scale) 4.28 (1.07) 3.85 (1.03) 2.12*
Sharing personal problems 2.15 (1.38) 2.08 (1.37) 0.26
Socializing after work 5.51 (1.57) 4.98 (1.62) 1.82
Exchanging confidences 5.67 (1.01) 5.30 (1.15) 1.93þ
Considering the other one to be a friend 3.64 (2.11) 3.67 (1.84) 0.09
Often going to lunch with each other 4.31 (1.89) 3.23 (1.73) 3.31***
Career development (scale) 4.98 (0.94) 5.31 (0.68) 2.28*
Mentor takes personal interest in mentee’s career 5.71 (1.24) 6.36 (0.60) 3.70***
Mentor placed mentee in important assignments 3.87 (1.59) 3.82 (1.47) 0.18
Mentor gives mentee special coaching on the job 4.97 (1.63) 5.21 (1.44) 0.88
Mentor advises mentee about promotional opportunities 4.03 (1.69) 4.82 (1.42) 2.78**
Mentor helps to coordinate professional goals 5.72 (1.13) 5.98 (0.83) 1.47
Mentor devoted special time and consideration to mentee’s career 5.57 (1.23) 5.72 (0.84) 0.77
Role modeling (scale) 5.62 (0.86) 4.98 (0.80) 4.25***
Mentee tries to model behavior after mentor 4.93 (1.22) 4.31 (1.10) 2.95**
Mentee admires mentor’s ability to motivate others 5.38 (1.32) 4.61 (0.88) 3.79***
Mentee respects mentor’s knowledge of the profession 6.39 (0.78) 5.66 (0.95) 4.70***
Mentee respects mentor’s ability to teach others 5.77 (1.10) 5.33 (1.01) 2.31*
þp<0.10;*p<0.05;**p<0.01;***p <0.001.
Note:The sample was n¼61 mentees and n ¼61 mentors with d.f.¼120.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
mentors more than the mentors thought.The t-tests on the item level reveal a mixed picture.Mentees
and mentors agreed the least on role modeling behaviors of the mentee,with all items showing
significant differences.For career development,the dyad partners significantly disagreed on two
behaviors:mentor advising mentee about promotional opportunities and mentor taking interest in
the mentee’s career.The other three means showed no significant differences,however.Psychosocial
support is the scale which mentors and mentees showed the least disagreement.Going to lunch quite
often was the only relationship behavior on which they disagreed.Overall,both the correlations and
the mean difference tests point to little correspondence between mentors’ and mentees’ perceptions of
their relationships.Examination of the data at both the scale level and the item level suggests that,if
there is any agreement,it is most likely regarding the psychosocial support behaviors.
Relations of mentoring,supervisory,and coworker
relationships with outcomes
The study’s other question regarded the potential relative effectiveness,for the mentee,of the relation-
ships between the mentee and the three other parties in the workplace:the formal mentor,the super-
visor,and the best-liked coworker.Table 3 shows the correlations between the relationships of the
mentee with these three parties and the three mentee outcomes:job satisfaction,organizational com-
mitment,and turnover intention.
Overall,it is apparent that the mentor relationships tended to have the least relationship with the
mentees’ outcomes.None of these nine correlations was significant.Both coworker and supervisor
relationships had a larger number of significant relations with the outcomes (for supervisors,eight
of the 12 correlations were significant at the p<0.05 level and another was marginally significant,
and for coworkers six of 12 were significant).Regarding relationships with supervisors,the subordi-
nate’s professional respect for the supervisor was strongly related to all three outcomes,and both affect
Table 3.Correlations of mentee outcomes with mentor,supervisor,and coworker relationships
Mentee outcomes
Job Organizational Turnover
satisfaction commitment intent
Mentor relationships
Career development 0.01 (60) 0.05 (61) 0.10 (61)
Psychosocial support 0.18 (59) 0.06 (60) 0.10 (60)
Role modeling 0.07 (60) 0.01 (61) 0.05 (61)
Supervisor relationships
Affect 0.26* (59) 0.05 (60) 0.26* (60)
Professional respect 0.53** (59) 0.27* (59) 0.44**(59)
Loyalty 0.49** (58) 0.15 (59) 0.26* (59)
Contribution 0.25þ (59) 0.29* (60) 0.01 (60)
Coworker relationships
Affect 0.21 (58) 0.44* (59) 0.17 (59)
Professional respect 0.43** (58) 0.12 (57) 0.18 (57)
Loyalty 0.28* (58) 0.33* (59) 0.27* (59)
Contribution 0.21 (56) 0.38** (57) 0.17 (57)
þp <0.10;*p <0.05;**p <0.01.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
and the perceived loyalty of the supervisor for the subordinate were related to two of the three
outcomes.Among the coworker relationships,loyalty of the coworker towards the employee was very
important,because it was related to all three outcomes.Overall,each coworker relationship was
related to at least one outcome.
Sixty-one per cent of the respondents were female.This was not representative of the company
workforce,because the mentoring programs had been implemented partly to address diversity issues
in both companies.The participants divided into 13 purely male mentoring dyads,20 purely female
mentoring dyads,25 dyads with female mentees and male mentors,and three dyads with male mentees
and female mentors.Running moderated regressions to check for possible gender differences in rela-
tionships between the mentoring behaviors and outcomes did not render any significant interactions.
As far as ethnicity differences are concerned,there were 32 purely Caucasian mentoring dyads,two
purely African–American dyads,and 27 mixed-ethnicity dyads in various constellations.Thus,this
study did not provide subsamples big enough to allow sound investigations on matching and non-
matching ethnicities.
The formal mentoring programs in these organizations were expected to result in the formal mentors
being important influences on the study’s three outcomes.Because the supervisory and coworker rela-
tionships in the organizations appeared to be stronger influences,we computed hierarchical regres-
sions to determine whether or not supervisor and coworker relationships actually had incremental
power (over the mentoring relationships) to predict the outcomes.In these regressions,all mentoring
relationships were entered on the first step and either (1) all supervisor relationships or (2) all coworker
relationships were entered on the second step.Table 4 shows that both supervisor and coworker rela-
tionships added incremental variance to the prediction of two of the three outcomes.Both supervisor
and coworker relationships predicted incremental variance in job satisfaction over the influence of
mentoring relationships.Supervisor relationships also had an incremental effect on turnover intention
but not on organizational commitment.Coworker relationships,on the other hand,made an incremen-
tal prediction for organizational commitment but not for turnover intention.The incremental predic-
tion of job satisfaction by supervisor relationships was particularly strong (R
change ¼0.47).Thus,
the relationships of employees with their supervisors and coworkers appear to be much more important
than their relationships with their formal mentors.
Table 4.Hierarachical regressions:effects of supervisor and coworker relationships while holding mentor
relationships constant
Criterion Predictors R
change Adjusted R
Job satisfaction Step one:Mentor relationships 0.01
Step two:Supervisor relationships 0.47** 0.41
Organizational commitment Step one:Mentor relationships 0.01
Step two:Supervisor relationships 0.12 0.02
Turnover intent Step one:Mentor relationships 0.02
Step two:Supervisor relationships 0.25** 0.17
Job satisfaction Step one:Mentor relationships 0.01
Step two:Coworker relationships 0.22* 0.12
Organizational commitment Step one:Mentor relationships 0.00
Step two:Coworker relationships 0.20* 0.09
Turnover intent Step one:Mentor relationships 0.05
Step two:Coworker relationships 0.11 0.03
*p <0.05;**p <0.01.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
Shared view of the mentoring relationship
There was no evidence that the mentoring relationships were perceived in the same way by the two
members of the mentor–mentee dyads.Even though well-established and behaviorally based measures
were used,it seems as if mentors and mentees can have a very different assessment of the relationship
they engage in.This is surprising if we account for the consistent time the mentors and mentees spent
together,which was 1.5 hours every 2 weeks on average,as well as their shared organizational culture
which should strengthen a common understanding.The organizational tenure,which could be used as
an indicator of the shared organizational culture,was 7 years for mentees and 17 years for mentors.
The tenure in the mentoring relationship was,by definition of course,the same for each person in the
dyad.Therefore,if a common organizational culture and common experience in the relationship would
help them to be closer and have a more consistent way of looking at things such as their mentoring
relationships,such effects were not apparent.
The observed discrepancies between mentors’ and mentees’ perceptions about their relationships
are especially critical if we consider ramifications such as (1) different beliefs and recommendations
of the mentor regarding career development options for the mentee or (2) embeddedness of a mentor-
ing relationship in a larger formal mentoring programwhich is steered by a programmanager who has
to intervene in the relationship in case of irregularities such as deterioration of trust,etc.The lack of
agreement between mentor and mentee perceptions of their relationship thus could be a signal of an
underdeveloped mentoring relationship.The prototypical mentor relationship is supposed to be very
close,which should foster common understanding and agreement about the relationship.It is rare to
look at mentor–mentee agreement about their own relationship,and so this study provides valuable
new information;the result is not encouraging,however.
It should be noted that similar lack of common perception in the supervisory domain was also men-
tioned by Gerstner and Day (1997),who reported a lack of leader–member agreement for the LMX
relationship in their meta-analysis of various LMX studies.Therefore,the lack of agreement about
reciprocal relationships across vertical organizational boundaries might be a common problem.Could
it be that the perceptions of people who occupy vertically different positions in the organizational hier-
archy are influenced by their inherent power differential?In particular,interactions between people of
very different statuses might be perceived by the parties in very different ways.The lower status person
can have a need to believe that he or she has made a favorable impression on the other person,and this
could get in the way of authenticity in the relationship.The upper status person might feel all is going
well when it is not,because the other person might not be willing to show weakness,disappointment,
or frustration.In addition,expectations of the two dyad partners regarding the mentoring behaviors the
other person should engage in for the relationship to become successful might be very different,and
this has been found to play an important role in the perceived exchange quality (Young & Perrewe
2000b).Because little previous research on mentoring had looked at the issue of congruence of per-
ceptions of the relationship,more research is appropriate in order to verify this finding.
Possible effects of the different relationships
on employee outcomes
One might expect that the mentoring and supervisor dimensions that most closely paralleled each
other would be related to the outcomes in similar ways,while this would not occur for mentoring
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
and supervisor dimensions that did not resemble each other.Similarities or differences in the content of
the relationships were not the key for interpreting the results,however.Rather than differences based
on dimensions,the obvious differences in results were due to the source of the relationship:mentors
versus supervisors or coworkers.Supervisor and coworker relationships were more important than
mentoring relationships in their potential effects on an individual’s organizational commitment,job
satisfaction and turnover intent.Mentoring did not play a significant role for any of the outcome
variables,which was a suprising result.This contradicts our expectations for mentoring relationships
based on social exchange theory,and it also is different from some previous findings,especially
regarding job satisfaction (Chao et al.,1992;Whitely & Coetsier,1993;Koberg,Boss,Chapell,&
We expected that turnover intentions and organizational commitment would be particularly suscep-
tible to influence by the social exchanges in the mentoring relationship in the current study.The men-
tors should have been obvious representatives of the organization,given their high organizational
status and the organizations’ sponsorship of the formal mentoring program.If the mentoring relation-
ships were,symbollically,exchange relationships with the organization,then the outcomes that repre-
sent continued exchange with the organization (turnover intention and organizational commitment)
should have been affected by the mentoring exchanges.
The results were consistent,however,with a more recent finding of Burke and McKeen (1997),of no
relationship between job satisfaction and career or psychosocial mentoring.The result could partly be
explained by the different amounts of time that mentees spend with their mentors,supervisors,and
coworkers that gives mentors simply much less time to shape their mentees’ attitudes in comparison
to supervisors or coworkers.If longer-termoutcomes had been measured,the mentors might have had
a stronger influence.
We must consider the obvious possibility,however,that formal mentoring programs of the type in
the present study are simply at a disadvantage for affecting mentee outcomes;mentors who are two
levels above the mentees are not normally in close contact with them,and because of their relatively
high level of responsibility,they might often be busy and not have a great deal of time to spend with a
mentee as an ‘extra’ task.There are two logical practical implications of this discussion.First,formal
mentor programs that involve mentors who are different from and usually above the mentees’ super-
visors in the organizational hierarchy should try to build in more time commitment to the mentoring
relationship.That is,the mentors and mentees should be required to spend more time together than the
minimum two hours per month in the formal mentoring programs in the present study.This might
make up for the apparent time deficit of mentors compared with supervisors and coworkers.On the
other hand,given that these upper-level mentors’ jobs did not naturally take them into close contact
with the mentees,and considering howbusy they were with their own work,it simply might be asking
a lot for themto spend even more time with the mentees.The second practical implication,therefore,is
that supervisors or coworkers should be chosen to performmentoring functions.They are likely to be
naturally closer (spatially) to the mentees during working hours,and they appeared to have some
advantage in terms of impact on the mentees in the present study.
It should also be noted that one of the earlier,more positive studies (Koberg et al.,1994) made no
distinction between supervisor and mentor as a counterpart of the mentee,and the supervisor effects
might therefore account for the mentor effects.A second earlier study (Chao et al.,1992) did inves-
tigate both formal and informal mentees’ perceptions but did not publish a specific result for formally
mentored individuals on job satisfaction.Instead,both formal and informal mentees were combined in
the analyses;this might be crucial,because formal mentees were only about a fifth of all mentees
investigated.Finally,a recent study by Ragins et al.(2000) found no differential effects of formal
and informal mentoring programs,although they did not distinguish between supervisors and others
who were mentors,a distinction that the present study suggests might be very important.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
As far as gender differences in mentoring dyads are concerned,previous literature has shown incon-
sistent results,and the present study did not find any moderator effects.Noe (1988) found females in
general to report significantly higher psychosocial benefits in the mentoring relationship than males,
and Goh (1991) found that women supervised by males tend to perceive less positive overall super-
visory mentoring behavior toward themselves compared to men supervised by males.In contrast to
that,Turban and Dougherty (1994) found no gender differences in received mentoring,a finding cor-
roborated by Ragins and Cotton (1999),who investigated matching and non-matching gender mentor-
ing dyads with regard to career development and psychosocial support and found no gender effects.
Therefore,we have no evidence that effectiveness of formal mentoring programs depends on the
genders of mentees and mentors,and the data presented in this study do point in the direction of no
existing gender differences.We also considered minority status as a possible moderator,but we did not
have a sample that would allow its empirical investigation.In general,research has shown that it is
more difficult for some minorities,especially African–Americans and Hispanics,to establish mentor-
ing relationships (Dreher & Cox,1996).At the same time,much of the research in this area refers to
informal relationships (Ragins,1997;Dreher &Cox,1996) or developmental relationships in a wider
sense (Greenhaus,Parasuraman,& Wormley,1990;Thomas,1990).More research in the future is
probably warranted on this issue.
This study’s distinction between a supervisor and a different,more formal mentor as the person
the respondent refers to when answering the survey is commonly ignored in many earlier studies
(e.g.,Noe,1988;Whitely &Coetsier,1993).This makes it all the more important that the present study
separated potential effects on the mentee that might come fromrelationships with multiple people who
could performmentoring-type roles.One must tentatively conclude at present that the effects of formal
mentoring programs on the mentee attitudes measured in the current study ( job satisfaction,turnover
intentions,and organizational commitment) are questionable,especially compared to potentially
similar but less formal mentoring-type relationships with others in the workplace.
Based on much of the applied literature on mentoring,one might logically develop formal mentor-
ing programs such as those in the present study,in which the mentor is not the supervisor but is higher
in the organizational hierarchy.It is possible that mentoring-type functions performed by the super-
visor (or even by coworkers) might be more effective than mentoring by someone else,however.It
might be a promising idea to have coworkers and supervisors take over mentoring functions,especially
when the goal is to foster job-related skill development (see also Eby,1997).We think this makes sense
for at least three reasons.First,although the present study purposely avoided examining mentoring
relationships with one’s supervisor or coworkers,this is consistent with the results,in which supervisor
and coworker relationships were more consistently related to mentee outcomes.Second,our examina-
tion of many prior studies found that supervisors were often mentors,and so this would not be unusual.
Third,Eby developed a model of mentoring in which lateral mentoring relationships are one dimens-
tion,and this fits with the concept of coworkers as mentors.The present research does not address
formal mentors who are supervisors (although much prior research does) or coworkers,but our results
suggest that Eby’s model might be useful for directing future research.
Still other previous findings help lead to a focus on supervisors versus mentors.Many studies dif-
fered in important features from the present one.In the study by Whitely,Dougherty,and Dreher
(1991),respondents did not participate in a formal mentoring program,and there was no distinction
between a mentor and supervisor as the dyadic counterpart of the mentee.Also,the mentoring received
was measured on the entire career span as opposed to a more specific time-limited relationship in this
study.Scandura and Schriesheim (1994) investigated supervisor–subordinate dyads in which the
amount of mentoring beyond LMX was measured.Also,only one company (and therefore one
mentoring culture) was investigated,while there were two in the present study.In Fagenson’s
(1989) study respondents indicated whether they had a mentor without making a clear distinction
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
between supervisor and mentor.This might be crucial,because research by Scandura and Schriesheim
(1994) shows that mentoring and LMX are not empirically different from the subordinate’s perspec-
tive,if both come from the same supervisor.Overall,previous research often did not clearly
distinguish between mentors and supervisors.Often they were the same person or could be the same
person.Therefore supervisory behaviors could not be distinguished frompurely mentoring behaviors.
The present study was the first to make this distinction clearly.Results suggest that mentoring func-
tions might be best performed by the supervisor rather than someone higher up in the hierarchy.
An important practical implication of the study,therefore,is that organizations should develop super-
visors rather than other upper-level managers as mentors.Further research is needed before this can be a
strong recommendation,however.Afirst limitation of the study is that the outcomes measured tended to
be short term,but a major goal of mentoring is long-term career development.From the present data,
one cannot be sure that the supervisor relationship had a stronger long-termeffect than the mentor rela-
tionship did.The outcomes in the present study were job satisfaction,organizational commitment,and
turnover intentions.The most future-oriented of these was turnover intention,which as measured here
was people’s intentions to leave the organization during the next year.It seems likely that effectiveness
in influencing long-termmentoring outcomes would require effects on short-termoutcomes as well,and
previous writers support that assumption that the study’s short-term outcomes should be affected by
mentoring (Aryee et al.,1996;Chao et al.,1992;Corzine et al.,1994;Fagenson,1989;Goh,1991;
Mobley et al.,1994;Ragins &Cotton,1999;Seibert,1999;Whitely &Coetsier,1993).Future research
should examine short-term outcomes such as those in the present study as potential mediating vari-
ables—mediating between mentoring activities and long-term outcomes such as career development.
A second limitation is that some of the measures had low reliabilities.One criterion (turnover
intent),one variable on the mentor questionnaire,and one variable fromeach of the three relationships
described on the mentee questionnaire (relationships with mentor,supervisor,and coworker) had
alphas that were only in the 0.60s.These reliabilities might have attenuated some of the empirical
relationships.This effect would be most pronounced in correlations for which both variables had
low reliabilities,i.e.,when mentor social support or supervisor contribution or coworker affect were
correlated with turnover intent.All three of these correlations were nonsignificant (Table 4).In sub-
sequent reanalyses,we corrected these three correlations for attenuation due to unreliablity and found
that only one of the corrected correlations reached statistical signficance with such corrections:affect
for the coworker (corrected r ¼0.26,p <0.05).In addition,the only previous findings we are aware of
regarding CMX and outcomes found generally weaker relationships between CMX (called CWX in
that study) and satisfaction and commitment (Sherony & Green,2002),again suggesting that the low
reliabilities in the present study may be a common problem in this area of research.It is noteworthy
that these results only reinforce the conclusion that the formal mentor is less likely to impact the men-
tee than are others in the workplace.In this case,the corrected coworker relationship reached statistical
significance,but the corrected mentor relationship did not.
Regarding the results for the coworker relationship,the study examined the single peer who was
closest to the focal person,but most people have multiple coworkers who can all interact with the per-
son either individually or as a group.It may be that the group of coworkers has a different impact,
perhaps a stronger one,than the single closest person.It remains for future research to determine
the potential mentoring function of coworkers as a group.
Another caveat refers to the cross-sectional data which were assessed fromone instrument adminis-
tered at one time.The nature of the data does not warrant strong causal inferences.An alternative
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
explanation for the relationship between supervisor relationship and job satisfaction would be,for
example,that employees with a higher satisfaction attribute more value to the supervisor relationship
and alter their perceptions of it.
Another possible explanation for the lack of results could stem from mentoring programs that did
not work out.During the data collection one of the authors visited the companies,and she talked to the
mentoring programmanagers and the consultants,and she participated in mentoring programmeetings
to talk about the study and ask for participation.During the time present at the companies there was not
the slightest hint that the programs might not work.The programmeetings were well attended and had
a friendly and cooperative atmosphere,and the participants (mentees and mentors) seemed as enthused
about the programs as the mentoring program managers and consultants were.Program managers
reported that it was part of their task to help monitor the ongoing relationships and to serve as a med-
iator when problems in a dyad arise in order to either help to correct them or to help to dissolve the
relationship if necessary.Only one of the programmanagers said she had experienced any (two) cases
during her work as a programmanager where it was necessary to dissolve a relationship—one of them
due to a break of confidence,the other due to unresolvable personality differences between the dyad
partners.Other than that,the programmanagers reported that the mentoring relationships worked well.
In addition,it is very probable that the response rate would have been much lower if the mentoring
relationships had not worked out.In that case it is likely that the participants would not have been very
motivated to fill out a survey about a relationship that is either a burden to them or practically non-
existent;they might have believed that filling out the surveys was a mere waste of time.
No study can practically measure all possible variables relevant to the three outcomes in the study
( job satisfaction,organizational commitment,and turnover intentions),and it is left to future research
to combine mentoring with other predictors in order to form a complete picture of these outcomes.
More importantly,we wonder specifically if mentoring combines with with other variables (e.g.,char-
acteristics of the situation).In other words,are there situational characteristics affecting the formal
mentoring-to-outcome relationships?If so,we then must ask whether there was something unique
about the current situation that would account for the results.The present study involved two firms,
and different firms can have a myriad of situational differences.When we analyzed the data holding
the firmconstant,the results were essentially the same,however,suggesting that differences in differ-
ent firms may not matter.
Furthermore,we came to a similar conclusion when analyzing the data holding time in the mentor-
ing relationship constant.Thus,we found little reason to believe that other variables made the study’s
results unique or unusual.An important contextual question is whether the formal mentoring programs
interfered with the formation of informal mentoring in the company.Because informal mentoring
practices might be more common organizational practices,it is possible that this happened in the
two companies at the time when the programs were implemented.The programs aimed at employees
that did not have a mentor at the time when they volunteered to participate.It seems possible that the
employees would have built up the same kind of relationships with some informal mentors if they had
not been supported through the formal mentoring programs.Informal mentoring relationships might
actually be more valuable,e.g.,because they might be more long-lasting and intense.In this case,and
referring to our outcome variables,supervisor and coworker relationships would do a better job.
Importance of LMX subqualities in the supervisor
and coworker relationship
Looking across the outcome variables,the supervisor relationships were especially consistently and
strongly related to job satisfaction,while the coworker relationships were especially related to
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
organizational commitment.Supervisor relationships had more and stronger (negative) relationships to
turnover intention.One interpretation is that low turnover intent is a formof individual repayment for
perceived organizational support and therefore might be a response to the relationship with the super-
visor,who might represent the organization for most employees.People might feel committed,on the
other hand,to the extent that they are tied in to their peers in the organization by having good relation-
ships with them.
For both the relationship to the supervisor as well as to the coworker,LMXsubqualities were used as
measures,and they were important for the outcomes in the study.This is noteworthy because LMX
was originally developed as a theory based on the reciprocity of behaviors.Some of the original beha-
viors in LMX theory relate closely to forms of performance between a supervisor and a subordinate
(Dansereau et al.,1975).The extension of LMX (and CMX in the present study) to include more
humanistic,interpersonal facets (Dienesch & Liden,1986;Liden & Maslyn,1998) appears to be an
important step in a promising direction,as all LMX and CMX factors appear to be potentially impor-
tant determinants of employee outcomes.Further research into CMXmay prove useful for understand-
ing coworker influences on employees.
Overall,supervisory and coworker relationships appear more important to the outcomes than formal
mentoring relationships do.One may look upon the one-to-one relationships with a supervisor or
coworker as either informal mentoring or as mentoring substitutes.Substitution may go both ways,
however.It is conceivable that mentoring (formal or informal) is primarily useful if other relationships
(e.g.,with supervisors or coworkers) in an organization are not functioning well.Research investigat-
ing multiple relationships’ potentials for mentoring functions has been suggested recently (Higgins &
Thomas,2001),and this suggestion is echoed by the present results.
The authors thank Rudolf Kerschreiter,Felix Brodbeck,Michael Frese,and Nina Gupta for their help-
ful comments.The data come froma thesis conducted by the first author and supervised by the second
author.Some of these data were presented at the 1999 conference of the American Psychological
Society in Denver.
Author biographies
Babette Raabe received her MA in Industrial/Organizational Psychology from Central Michigan
University and her German Diploma of Psychology from the University of Hamburg.She is currently
working in personnel development for Siemens Qualification and Training in Munich and teaches
career development at the Universities of Giessen and Munich.Her research interests include career
development,mentoring,LMX,self-management,training evaluation and social networks.
Terry A.Beehr received his PhDfromthe University of Michigan in (Organizational) Psychology.He
is currently a Professor and Director of the PhD Programin Industrial and Organizational Psychology
at Central Michigan University.His research interests include careers,mentoring and employee devel-
opment,occupational stress,retirement and workplace social support.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
Anderson,N.R.,& West,M.A.(1998).Measuring climate for work group innovation:development and
validation of the team climate inventory.Journal of Organizational Behavior,19,235–258.
Aryee,S.,Chay,Y.W.,& Chew,J.(1996).The motivation to mentor among managerial employees.Group and
Organization Management,21,261–277.
Atwater,L.E.,& Yammarino,F.J.(1992).Does self-other agreement on leadership perceptions moderate the
validity of leadership and performance predictions?Personnel Psychology,45,141–164.
Blake,R.R.,& Mouton,J.S.(1982).Management by grid principles or situationalism:which?Group and
Organization Studies,7,207–210.
Blau,P.M.(1964).Exchange and power in social life.New York:Wiley.
Burke,R.J.,& McKeen,C.A.(1997).Benefits of mentoring relationships among managerial and professional
women:a cautionary tale.Journal of Vocational Behavior,51,43–57.
Camman,C.,Fichman,M.,Jenkins,D.,& Klesh,J.(1979).The Michigan Organizational Assessment
Questionnaire.Unpublished Manuscript,University of Michigan,Ann Arbor.
Chao,G.T.,Walz,P.M.,&Gardner,P.D.(1992).Formal and informal mentorships:a comparison on mentoring
functions and contrast with nonmentored counterparts.Personnel Psychology,45,619–636.
Corzine,J.B.,Buntzman,G.F.,& Busch,E.T.(1994).Mentoring,downsizing,gender and career outcomes.
Journal of Social Behavior and Personality,9,517–528.
Cropanzano,R.,Rupp,D.E.,Mohler,C.J.,& Schminke,M.(2001).The roads to organizational justice.In
G.Ferris (Ed.),Research in personnel and human resources management (pp.1–113).Oxford:Elsevier
Dansereau,F.,Graen,G.,& Haga,W.J.(1975).A vertical dyad linkage approach to leadership within formal
organizations.Organizational Behavior and Human Performance,13,46–78.
Dienesch,R.M.,&Liden,B.(1986).Leader–member exchange of leadership:a critique and further development.
Academy of Management Review,11,618–634.
Dreher,G.F.,& Cox,T.H.(1996).Race,gender and opportunity:a study of compensation attainment and the
establishment of mentoring relationships.Journal of Applied Psychology,81(3),297–308.
Eby,L.T.(1997).Alternative forms of mentoring in changing organizational environments:a conceptual
extension of the mentoring literature.Journal of Vocational Behavior,51,125–144.
Fagenson,E.A.(1989).The mentor advantage:perceived career/job experiences of prote
s versus non-prote
Journal of Organizational Behavior,10,309–320.
Fleishman,E.A.,& Harris,E.F.(1962).Patterns of leadership behavior related to employee grievances and
turnover.Personnel Psychology,15,43–56.
Gerstner,C.R.,& Day,D.V.(1997).Meta-analytic review on leader–member exchange theory:correlates and
construct issues.Journal of Applied Psychology,82,827–844.
Godshalk,V.M.,&Sosik,J.J.(2000).Does mentor–prote
agreement on mentor leadership behavior influence
the quality of a mentoring relationship?Group & Organization Management,25,291–317.
Goh,S.C.(1991).Sex differences in perceptions of interpersonal work style,career emphasis,supervisory
mentoring behavior,and job satisfaction.Sex Roles,24,701–710.
Greenhaus,J.H.,Parasuraman,S.,& Wormley,W.M.(1990).Effects of race on organizational experiences,job
performance evaluations,and career outcomes.Academy of Management Journal,33,64–86.
Griffeth,R.W.,Hom,P.W.,& Gaertner,S.(2000).A meta-analysis of antecedents and correlates of employee
turnover:update,moderator tests,and research implications for the next millennium.Journal of Management,
Graen,G.B.,&Uhl-Bien,M.(1998).Relationship-based approach to leadership:development of leader–member
exchange (LMX) theory of leadership over 25 years:applying a multi-level multi-domain perspective.In F.
Dansereau,& F.J.Yammarino (Eds.),Leadership:The multiple-level approaches:Contemporary and
alternative (pp.103–155).Stamford,CN:JAI Press.
Hackman,J.R.(1992).Group influences in organizations on individuals.In M.D.Dunnette,& L.M.Hough
(Eds.),Handbook of industrial and organizational psychology (Vol.3,2nd edn,pp.199–267).Palo Alto,CA:
Consulting Psychologists Press.
,R.,& Secord,P.F.(1972).The explanation of social behavior.Oxford:Blackwell.
Higgins,M.C.,& Kram,K.E.(2001).Reconceptualizing mentoring at work:a developmental network
perspective.Academy of Management Review,26,264–288.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
Higgins,M.C.,&Thomas,D.A.(2001).Constellations and careers:toward understanding the effects of multiple
developmental relationships.Journal of Organizational Behavior,22,223–247.
Homans,C.G.(1958).Social behavior as exchange.American Journal of Sociology,63,597–606.
Jones,E.E.,& Gerard,H.B.(1967).Foundations of social psychology.New York:Wiley.
Koberg,C.S.,Boss,R.W.,Chapell,D.,& Ringer,R.C.(1994).Correlates and consequences of protege
mentoring in a large hospital.Group and Organization Management,19,219–239.
Kram,K.E.(1983).Phases of the mentor relationship.Academy of Management Journal,26,608–625.
Kram,K.E.(1985).Mentoring at work:Developmental relationships in organizational life.Glenview,IL:Scott,
Liden,R.C.,&Maslyn,J.M.(1998).Multidimensionality of leader–member exchange:an empirical assessment
through scale development.Journal of Management,24,43–72.
Major,D.,Kozlowski,S.W.J.,Chao,G.T.,& Gardner,P.D.(1995).A longitudinal investigation of newcomer
expectations,early socialization outcomes,and the moderating effects of role development factors.Journal of
Applied Psychology,80,418–431.
Maurer,T.J.,&Palmer,J.K.(1999).Management development intentions following feedback:role of perceived
outcomes,social pressures,and control.Journal of Management Development,18,733–751.
Maurer,T.J.,Pierce,H.R.,& Shore,L.M.(2002).Perceived beneficiary of employee development activity:a
three-dimensional social exchange model.Academy of Management Review,27,432–444.
McManus,S.E.,& Russell,J.E.A.(1997).New directions for mentoring research:an examination of related
constructs.Journal of Vocational Behavior,51,145–161.
Mobley,G.M.,Jaret,C.,Marsh,K.,& Lim,Y.Y.(1994).Mentoring,job satisfaction,gender,and the legal
profession.Sex Roles,31,79–98.
Mowday,R.T.,Steers,R.,& Porter,L.(1979).The measurement of organizational commitment.Journal of
Vocational Behavior,30,143–158.
Mullen,E.J.(1994).Framing the mentoring relationship as an information exchange.Human Resource
Management Review,4,257–281.
Nadig,V.(1994).The measurement of multiple commitments:scale development and validation.Dissertation
Abstracts International,55,05B (University Microfilms No.AAI95–27838).
Noe,R.A.(1988).An investigation of the determinants of successful assigned mentoring relationships.Personnel
Phillips-Jones,L.(1982).Mentors and prote
s:How to establish,strengthen,and get the most from mentor–
relationships.New York:Arbor House.
Podsakoff,P.M.,Mackenzie,S.B.,& Ahearne,M.(1997).Moderating effects of goal acceptance on the
relationship between group cohesiveness and productivity.Journal of Applied Psychology,82,974–983.
Ragins,B.R.(1997).Antecedents of diversified mentoring relationships.Journal of Vocational Behavior,51,
Ragins,B.R.,&Cotton,J.L.(1999).Mentor functions and outcomes:a comparison of men and women in formal
and informal mentoring relationships.Journal of Applied Psychology,84,529–550.
Ragins,B.,&Scandura,T.A.(1999).Burden or blessing?Expected costs and benefits of being a mentor.Journal
of Organizational Behavior,20,493–509.
Ragins,B.R.,Cotton,J.L.,& Miller,J.S.(2000).Marginal mentoring:the effects of type of mentor,quality of
relations,and programdesign on work and career attitudes.Academy of Management Journal,43,1177–1194.
Revicki,D.A.,Whitley,T.W.,&Gallery,M.E.(1993).Organizational characteristics,perceived work stress,and
depression in emergency medicine residents.Behavioral Medicine,19,74–81.
Riley,S.,& Wrench,D.(1985).Mentoring among female lawyers.Journal of Applied Social Psychology,15,
Riordan,C.M.,& Griffeth,R.W.(1995).The opportunity for friendship in the workplace:an underexplored
construct.Journal of Business and Psychology,10,141–154.
Russell,J.E.A.,& Adams,D.M.(1997).The changing nature of mentoring in organizations:an introduction to
the special issue on mentoring in organizations.Journal of Vocational Behavior,51,1–14.
Scandura,T.A.(1998).Dysfunctional mentoringrelationships andoutcomes.Journal of Management,24,449–467.
Scandura,T.A.,& Ragins,B.R.(1993).The effects of sex and gender role orientation on mentorship in male-
dominated occupations.Journal of Vocational Behavior,43,251–265.
Scandura,T.A.,& Schriesheim,C.A.(1994).Leader–member exchange and supervisor career mentoring as
complementary constructs in leadership research.Academy of Management Journal,37,1588–1602.
Seashore,S.E.,Lawler,E.E.,Mirvis,P.H.,Cammann,C.(Eds.).(1982).Assessing organizational change:A
guide to methods,measures,and practices.New York:Wiley.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
Seibert,S.(1999).The effectiveness of facilitated mentoring:a longitudinal quasi-experiment.Journal of
Vocational Behavior,54,483–502.
Sherony,K.M.,& Green,S.G.(2002).Coworker exchange:relationships between coworkers,leader–member
exchange,and work attitudes.Journal of Applied Psychology,87,542–548.
Tepper,B.J.(1995).Upward maintenance tactics in supervisory mentoring and nonmentoring relationships.
Academy of Management Journal,38,1191–1205.
Thibaut,J.W.,& Kelley,H.H.(1959).The social psychology of groups.New York:Wiley.
Thomas,D.A.(1990).The impact of race on managers’ experiences of developmental relationships (mentoring
and sponsorship):an intra-organizational study.Journal of Organizational Behavior,11,479–492.
Thomas,D.A.(1993).Racial dynamics in cross-race developmental relationships.Administrative Science
Turban,D.B.,&Dougherty,T.W.(1994).Role of prote
personality in receipt of mentoring and career success.
Academy of Management Journal,37(3),688–702.
Tyler,K.(1998).Mentoring programs link employees and experienced execs.HR Magazine,43(5),98–103.
Whitely,W.T.,&Coetsier,P.(1993).The relationship of career mentoring to early career outcomes.Organization
Whitely,W.,Dougherty,T.W.,& Dreher,G.F.(1991).Relationship of career mentoring and socioeconomic
origin to managers’ and professionals’ early career success.Academy of Management Journal,34,331–351.
Wohlers,A.J.,& London,M.(1989).Ratings of managerial characteristics:evaluation difficulty,co-worker
agreement,and self-awareness.Personnel Psychology,42,235–261.
Young,A.M.,& Perrewe
,P.L.(2000a).The exchange relationship between mentors and proteges:the
development of a framework.Human Resource Management Review,10,177–209.
Young,A.M.,& Perrewe
,P.L.(2000b).What did you expect?An examination of career-related support and
social support among mentors and prote
s.Journal of Management,26,611–632.
Yukl,G.,& Van Fleet,D.D.(1992).Theory and research on leadership in organizations.In M.D.Dunnette,&
L.M.Hough (Eds.),Handbook of industrial and organizational psychology (Vol.3,2nd edn pp.147–197).
Palo Alto,CA:Consulting Psychologists Press.
Zacarro,S.J.,& Dobbins,G.H.(1989).Contrasting group organizational commitment:evidence for differences
among multilevel attachments.Journal of Organizational Behavior,10,267–273.
Zohar,D.(2000).A group-level model of safety climate:testing the effect of group climate on microaccidents in
manufacturing jobs.Journal of Applied Psychology,85,587–596.
Appendix:Items for the Mentoring and LMX
Mentoring items (after Scandura and Ragins,1993)
Psychosocial support
I share personal problems with my mentor.
I socialize with my mentor after work.
I exchange confidences with my mentor.
I consider my mentor to be a friend.
I often go to lunch with my mentor.
Career development
My mentor takes a personal interest in my career.
My mentor has placed me in important assignments.
My mentor gives me special coaching on the job.
Note that the items in the coworker scales directly parallel the LMX items in the supervisors scale.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)
My mentor advised me about promotional opportunities.
My mentor helps me to coordinate my professional goals.
My mentor has devoted special time and consideration to my career.
Role modeling
I try to model my behavior after my mentor.
I admire my mentor’s ability to motivate others.
I respect my mentor’s knowledge of the profession.
I respect my mentor’s ability to teach others.
LMX supervisor items (after Liden and Maslyn,1998)
I like my supervisor very much as a person.
My supervisor is a lot of fun to work with.
My supervisor is the kind of person one would like to have as a friend.
Professional respect
I respect my supervisor’s knowledge of and competence on the job.
I am impressed with my supervisor’s knowledge of his/her job.
I admire my supervisor’s professional skills.
My supervisor defends my work actions to a superior,even without complete knowledge for the issue
in question.
My supervisor would defend me to others in the organization if I made an honest mistake.
My supervisor would come to my defense if I were ‘attacked’ by others.
My supervisor can depend on me when we are overloaded with work.
I am willing to apply extra efforts,beyond those normally required,to meet my supervisor’s work
I do not mind working hardest for my supervisor.
I do work for my supervisor that goes beyond what is specified in my job description.
Copyright#2003 John Wiley & Sons,Ltd.J.Organiz.Behav.24,271–293 (2003)