Pittsburg State University Journal of Managerial Issues

Subordinate Self-esteem and Abusive Supervision
Author(s): James P. Burton and Jenny M. Hoobler

Source: Journal of Managerial Issues, Vol. 18, No. 3 (Fall 2006), pp. 340-355
Published by: Pittsburg State University
Stable URL: http://www.jstor.org/stable/40604544
Accessed: 12-03-2017 10:55 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
Pittsburg State University is collaborating with JSTOR to digitize, preserve and extend access to Journal
of Managerial Issues
This content downloaded from 111.68.97.226 on Sun, 12 Mar 2017 10:55:17 UTC
All use subject to http://about.jstor.org/terms
JOURNAL OF MANAGERIAL ISSUES
Vol. XVIII Number 3 Fall 2006: 340-355
Subordinate Self-esteem and Abusive Supervision*
James P. Burton
Assistant Professor oj Business
University of Washington, Bothell
Jenny M. Hoobler
Assistant Professor of Management
University of Illinois at Chicago
While physical acts of workplace vi-healthy workforce. Patterns of nega-

olence have received much attention tive interaction can be dangerous in
in the popular press in recent years,
that they threaten to damage organi-
academic researchers have begun in- zations, careers, and people (Masuch,
vestigating lesser forms of workplace
1985). In our study we tested hypoth-
mistreatment (e.g., Neuman and eses designed to uncover deleterious
Baron, 1997). While less sensationalconsequences for subordinates when
perhaps, verbal and passive forms they
of perceive their bosses' behavior
to be abusive. A handful of studies
aggression, such as yelling, bullying,
and humiliation, not only happen
have investigated the consequences
more frequently than active violence
of abusive behavior in managers. Tep-
per and colleagues (2000, 2004)
but can also be extremely damaging,
found that abusive supervisors engen-
contributing ultimately to workplace
stress and target demoralization
dered subordinate turnover, work-
(Keashly, 1998). Some studies suggest
family conflict, job and life dissatisfac-
that workplace violence occurs in tion and psychological distress.
20% of workplaces (Romano, 1994), Congruently, Duffy and Ferrier
(2003) found that abusive supervi-
yet almost twice that many workplaces
are the site of more subtle, verbally
sion was a significant predictor of sub-
harassing behavior or thoughtless, ordinate distrust and low organiza-
tional commitment. Hence, we
negative acts (Bjorkqvist et al, 1994).
Understanding the ramifications ofidentified abusive supervision as a
workplace aggression is important workplace
in Stressor that has serious
the quest to foster an emotionally implications for organizational mem-
* The authors wish to thank Terry Mitchell and Tom Lee for their helpful comments on p
versions of this manuscript. Both authors contributed equally to the article.
JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006
(340)
Subordinate Self-esteem and Abusive Supervision 341
colored
bers and, in particular, abused sub-by the status, power, and re-
ordinates. source differentials common to that
The negative consequences of abu- relationship. Because bosses are com-
sive supervision on subordinate emo- monly the gatekeepers to employee
tional well-being have frequentlyadvancement, compensation, and
been alluded to in previous work feedback, when this relationship is a
(e.g., Tepper, 2000), but have not dysfunctional one, it stands to have
been tested conclusively in the form particularly salient and devastating
of low subordinate self-esteem. This
consequences for employees. Early
research seeks to underscore the re-
evidence points to abused subordi-
nates experiencing greater psycho-
lationship between supervisors' abu-
sive behavior and subordinate self-es-
logical distress, job and life dissatis-
teem and to explore whether genderfaction, and intentions to quit their
influences this relationship. jobs, as compared to non-abused col-
In the sections that follow, we pro- leagues (Keashly et al, 1994). Ash-
vide some background on abusive su- forth (1997) found that a more gen-
pervision and then link this to em-eral type of antisocial organizational
ployee self-esteem. We then describebehavior, petty tyranny, was positively
the study we undertook to examine related to subordinate frustration, re-
the relationship between abusive su-actance, helplessness, and work alien-
pervision and subordinate self-esteem
ation, and negatively related to work
and discuss implications for research
unit cohesiveness, work performance,
and practice. and leader endorsement.
THEORY AND BACKGROUND Abusive Supervision and

State Self-esteem
Abusive Supervision
Historically, the majority of research
Abusive supervision has evolved
in self-esteem has examined one's gen-
from earlier, related constructs such eral or global level of self-esteem and
as petty tyranny (Ashforth, 1997) and all but ignored the role that temporary
non-physical workplace aggression affective feelings regarding one's self
(Neuman and Baron, 1997). This may have on a variety of behavior
construct has recently been defined (Crocker and Wolfe, 2001). Recent
as the ' 'display of hostile verbal and work, however, addressed the role that
non-verbal behaviors, excluding phys- a person's state self-esteem plays in be-
ical contact" (Tepper, 2000: 178). It havior (e.g., Vermunt et al, 2001).
is a subjective assessment made by State self-esteem (Heatherton and Po-
employees regarding their supervi- livy, 1991) refers to momentary
sor's behavior toward them. Exam- changes in a person's level of self-es-
ples of abusive supervision includeteema in response to some situational
boss demeaning, belittling, under- stimulus. Heatherton and Polivy dem-
mining, or taking credit for the work onstrated that a person's self-esteem
of a subordinate. These behaviors re- fluctuates around their general base-
flect indifference, hostility, and often-
line level of self-esteem. These tempo-
times deviance (Tepper, 2000) . rary fluctuations can be positive or
Abusive supervision is specific tonegative depending on what the per-
the supervisor-subordinate dyad and son encounters in their daily life (e.g.,
342 Burton and Hoobler
positive feedback, counters

failure). an abusiveInsupervisor,
addition, this
communicates
they found that state information regarding
self-esteem is dis-
their social
tinct from a person's standing in the group and
mood.
The research evidence suggests
leads to temporary that
reductions in state
self-esteem.is more affected
a person's self-esteem
by negative rather Although
thanpast research has notevents
positive di-
(e.g., Schroth and Shah,
rectly examined the 2000).relationship Inbe- ad-
dition, individuals are much more tween abusive supervision and self-es-
likely to remember negative interac- teem, Ashforth (1997) found that
tions with their supervisors and recall abusive workplace behavior, which he
these negative events with intense operationalized as petty tyranny, was
emotion (Dasborough and Ashkanasy, positively related to frustration, stress,
2003). Therefore, it is likely that one reactance, helplessness, and work al-
type of negative event that may cause ienation, and negatively related to
fluctuations in a person's state self-es- leader endorsement, work unit cohe-
teem is an episode of abusive supervi- siveness, and performance for subor-
sion. Examining the impact of abusive dinates. Keashly, Trott, and McLean
supervision on self-esteem is important (1994) established, not surprisingly,
because self-esteem can be highly cor- that employees who experienced abu-
related with job performance, as re- sive supervision also experienced less
vealed in a recent meta-analysis (Judge job satisfaction. Finally, research in or-
and Bono, 2001). If a person's self-es- ganizational-based self-esteem demon-
teem is affected by an episode of abu- strated that feelings of self-worth are
sive supervision, this may in turn have influenced by how people feel they
negative consequences for the organi- have been treated (e.g., McAllister and
zation as a whole.
Bigley, 2002). Therefore, we believe
Cooley (1902) observed that a per- that an individual's temporary feelings
son's self-view is highly contingent on of self-worth will be negatively related
how they believe they are perceived by to abusive supervision.
others. Involvement in a relationship Hypothesis 1: Abusive supervision will be nega-
that is characterized by a high degree tively related to subordinates state self-esteem.
of trust and respect communicates a
sense of communality and being held While we expect abusive supervision
in high regard (Brockner, 2002). On to have universally deleterious out-
the other hand, a lack of trust and re- comes for targets, these effects may be
spect while interacting with others can particularly pronounced for specific
individuals. Theoretical work on the
lead to feelings of exclusion and a loss
of social identity (Lind et al, 2001). In- psychology of the self-concept (how in-
clusion in exchange relationships dividuals perceive themselves) posits
(such as with a supervisor) satisfies athat the three sources of information
person's social or psychological need by which humans come to view and
for self-esteem and affiliation (Brock- evaluate themselves are: reflected ap-
ner, 2002). Furthermore, failure to praisals (reactions of others to us
gain acceptance or inclusion in inter-through which we learn self-relevant
personal situations has been shown to information), social comparisons
lead to reductions in one's self-esteem (comparing ourselves to others in or-
(Leary et al, 1995). Consistent with this der to generate self-evaluative infor-
logic, we expect that when a person en- mation), and self-perceptions (obser-
vations of and attributions for our own state self-esteem such that females will experience
behavior as a source of self-evaluative lower levels of state self-esteem following an abu-
sive supervision episode.
information) (Gecas, 1982). Gender role
socialization theory posits that males are
socialized to focus on self-esteem via METHOD
the ' 'achievement domain," whereas
females emphasize self-esteem derived In this study, we used written sce-
from the interpersonal domain (Whit- narios to manipulate a situation tha
ley, 1983). Therefore, men get infor- could occur in the workplace and
mation about themselves from theirthat should cause perceptions of abu
self-perceptions and social compari-
sive supervision. The subjects were in
sons, while women are more depend-structed to "play the part" of the per-
ent upon reflected appraisals son in the scenario. Greenberg an
(Schwalbe and Staples, 1991). While Eskew (1993) state that this is an ef
men form their self-concept through fective method for gauging how
replaying and interpreting their past someone would react to a similar sit-
behaviors and sizing themselves up as uation in an organization. In addi-
compared to referent others, women, tion, Wiseman and Levin (1996)
on the other hand, seek out informa- found that individuals make similar
tion regarding others' opinions of choices whether they are in hypo-
them. Women, in that their self-esteem thetical or "real" situations. We be-
is based on reflected appraisals, dem- lieve this method is appropriate for
onstrate a higher need for and sensi- testing our hypotheses since abusive
tivity to social approval. Because supervision is a fairly new addition to
women have learned to value sociabil- the organizational behavior field.
Therefore, highly controlled, inter-
ity and smooth-functioning relation-
ships (Schwalbe and Staples, 1991),nally valid studies are needed that can
demonstrate a clear link between an
when they encounter a dysfunctional
relationship, such as an abusive boss, itabusive supervision episode and a
particular outcome (McGrath, 1964).
is particularly detrimental to their self-
concept and their self-esteem. For Current human subject review pro-
those who rely on reflected apprais-cedures make it unlikely research
als (more likely women), self-esteem
subjects would be allowed to directly
tends to develop from a boss's ex-
experience manipulated abusive su-
pression of liking, approval, compe-
pervision. Therefore, scenarios may
tence, and worth. When a boss inbe the best method to test this phe-
stead exhibits antisocial behavior nomenon in controlled settings. Fi-
toward a subordinate, a man may nally,
en- the directionality of the rela-
gage self-evaluative information tionship between abusive supervision
from sources other than his boss, and state self-esteem could be diffi-
but a woman is more likely to expe- cult to interpret in a correlational,
rience decrements in self-esteem field study. Specifically, individuals
given that her boss is the designated with low self-esteem may be more sen-
organizational source of perform- sitive to the treatment they receive
ance appraisal and the embodiment from their supervisor due to a fragile
of organizational authority. self-concept and greater sensitivity to
social cues (Brockner, 1988). There-
Hypothesis 2: Subordinate gender will moderate
fore,
the relationship between abusive supervision and individuals with low self-esteem
may be more likely ratoryto session scheduled

rate during
supervisors
as abusive. In addition, self-esteem phase 2.
has been shown to be related to per- In order to ensure the confidenti-
formance (Judge and Bono, 2001); ality and anonymity of the partici-
therefore, individuals with lower self- pants' answers, subjects did not re-
esteem may be more likely to receive port their names or student ID
negative feedback, close monitoring numbers. To match the surveys from
and so forth from their supervisors the two different time periods, we
due to their lower levels of perform- adopted the approach used by Fedor
ance. Using a carefully controlled set- and his colleagues (2001). Specifi-
ting allows us to discount these alter- cally, a series of "identifier" ques-
native explanations and establish the tions were used to match the various
directionality between abusive super- surveys (e.g., "add up the last two
vision and state self-esteem. numbers of your house or apart-
ment's street address"). This method
Procedures was very successful in this study. Out
of the 196 participants in the study,
Phase 1. Subjects were recruited only four responses were unable to be
from introductory management clas- matched to a specific person.
Phase 2. Two to three weeks after
ses in the business school of a large
western university. During phase 1, participants completed the mea-
the
the experimenter went to each class sure of global self-esteem, they partic-
ipated in a brief laboratory session to
that was offering extra credit for their
gauge their reaction to a scenario in-
students' participation in this study.
volving a supervisor-subordinate in-
The potential subjects were told that
the researchers were interested in teraction. The scenario represents
behaviors indicative of abusive super-
how a person's mood influences their
visory behavior. That is, according to
behavior in work settings. In addi-
the content domain illustrated in
tion, all participants were guaranteed
Tepper's (2000) scale, the scenario
that the information we received
details behaviors on the part of a su-
from them was completely anony-
pervisor
mous and that there were no right or designed to be interpreted as
wrong answers. For those students rude, deceptive, humiliating, and un-
who agreed to participate, theyfair.1 im-
mediately completed a global self-es- When the participants entered the
teem measure. At this time, students room, they received a booklet with a
also completed various demographic cover sheet titled "Supervisor-Subor-
measures and signed up for a labo- dinate Relations in Organizational
1 To check whether our scenario adequately reflected the content domain of abusive supervision, we
sampled 60 members of an undergraduate human resource management course at a Midwestern
university. These students read and heard the scenario. Afterward, individually, they rated the
extent to which the scenario reflected a subset of Tepper's (2000) scale of abusive supervision.
Using a five-point scale (1 = strongly disagree; 5 = strongly agree), the subjects were asked to
report the degree to which the events of the scenario mirrored: ridicule, telling an employee his
or her thoughts or feelings are stupid, not giving an employee credit for a job or an idea requiring
a lot of effort, being rude to an employee, and lying to an employee. The student responses to the
scenario (mean = 4.12) indicate that the scenario accurately reflected an episode of abusive su-
pervision.
script involves
Settings." The subjects were asked to the conversation you had
with your supervisor:
complete the four ' 'identifier" ques-
tions and to not open the booklet un-
"I just looked at the recommendation
you want me to send to my superiors. I
til instructed by the experimenter.
don't know what you are thinking. If I
When everyone in the room had com-
send this suggestion to my supervisors,
pleted the identifier questions, it
they
will make me look bad in front of
were instructed to open the booklet
them. People at this organization expect
to the first page. In addition, the supervisors
sub- to think of these things and
jects were instructed to assumenotthat a simple counter clerk."
they were the employee beingYoudis- try to make a comment, but the boss cuts
cussed in the text. At this point, the
you off
participants read the first part ofHethe

says, "Any type of this behavior in
scenario. the future will be severely reprimanded.
You have worked in your current position at the
In fact, you may lose your job."
university for the past 2 years. Specifically, you You again try to make your point, but again
work for one of the cafés located throughout the are interrupted.
campus in academic buildings where students,
faculty, and administrators can grab a quick bite Your boss says, "Stop acting like a su-
to eat, get something to drink, or simply meet and pervisor and focus on your own job.
chat. Your duties include serving customers a va- Now, get back to work."
riety of food and drinL· (e.g., coffee, soup, sand-
wiches) and collecting payment for these items.To make matters worse, two weeks af-
Your café has a suggestions-award program. Theter your "conversation" with your
café encourages its student employees to submitboss, you learn that he has taken your
suggestions to improve workplace procedures.
idea and presented it to his superiors
Currently, your café maintains a separate office as "his own idea." Your boss is now
where it stores the materials it needs to serve its
receiving praise from his superiors re-
customers (e.g., coffee cups, soup bowls, napkins).
Recently, you submitted a suggestion to your su-
garding his bright ideas and contri-
pervisor that may reduce the current level of sup-bution to the café and the university.
plies on hand in your café. Specifically, you rec-
You learn that the idea will save the
ommend that the department adopt a just-in-time café thousands of dollars per month.
delivery schedule for its inventory. Your sugges-
In the second condition, subjects
tion could save your business a lot of money by
cutting the cost of maintaining inventory. You
simply read that their suggestion was
went out of your way and checked with suppliersapproved for implementation. It read
and were assured by them that they could supplyas follows:
the necessary materials to the café within 48 hours
Two weeL· after you make your suggestion to your
of an order.
supervisor, you learn that your idea has been ap-
proved for implementation. Your boss thanks you
The subjects were randomly as- for your suggestion.
signed to one of two conditions. In

Following the participants' reading
the first condition, students read a
of the scenario, they answered a series
written "transcript" of a conversation
of questions examining their reaction
where they were reprimanded for
to the particular scenario. Specifi-
making the suggestion. It read as fol-
lows:
cally, subjects responded to items de-
signed to measure the effect of the
Upon reading your suggestions, your boss scenario on their current level of self-
becomes angry. In front of your co-workers
and visibly upset, he asks to speak to you in
esteem. Following their completion
private. Once you enter his office, he se- of this activity, the participants were
verely criticizes you. The following tran- debriefed and excused.
Participants moment," having just read the sce-

nario. Sample items include "I feel
One hundred ninety-six individuals that others respect and admire me"
agreed to participate in the study and and "I feel good about myself." Sub-
completed the first measure of self- jects were instructed to circle the ap-
esteem and various demographic var- propriate "dot" on a ten-point scale.
iables. Of these individuals, 49.2% Although the Heatherton and Po-
were female and 49.7% were male livy scale is one of the most frequently
(this does not quite sum to 100% used measures of state self-esteem, it
since not all participants indicatedhas never been tested using confirm-
their gender). The participants' av- atory factor analysis techniques. This
erage age is 22.14 (SD = 4.74). is especially surprising since many of
Ninety-six percent of the participants the items reported in Heatherton
were business school students, while and Polivy (1991) had significant
the remaining seven subjects had ma- cross loadings across three factors
jors outside the business school (e.g., (appearance, social, and perform-
psychology, engineering) . Ninety- ance). In addition, some of the items
seven percent of the subjects were appeared to have been placed in a
juniors or seniors and the subjects av- category strictly on the results of the
eraged 5.6 years of work experience. exploratory factor analysis. For ex-
The study was completely voluntary ample, the item "I feel good about
and some participants removed myself loaded on the "appearance"
themselves from the remaining seg- factor while "I feel displeased with
ments of the study or inadvertently myself (reverse coded) loaded on
missed the laboratory section of the the "social" factor. Therefore, we be-
study. Forty-four subjects were unable lieve it is time to subject this test to a
to complete all phases of the study, more rigorous factor analytic ap-
yielding a sample size of 152. There proach.
were no significant differences be- Based on the items we selected for
tween those subjects who completed this study (we chose not to select
the entire study and those who did items referring to "appearance"
not for self-esteem level or any of the since we did not believe this to be rel-
demographic variables. evant for our study) , we first tested a
one-factor solution. The Chi-square
Measures statistic (206.84) was significant, but
this was not unexpected based on our
sample size. We relied on several
State Self-esteem. Each person's cur-
rent state of self-esteem was measuredgoodness-of-fit indices to assess the
using items modified from Heather- overall model fit. The results indicate
ton and Polivy (1991). This scale has that a one-factor solution fits the data
strong internal consistency and dis- well (CFI = .97, TLI = .96, IFI = .97,
tinguishes global self-esteem and NFI = .96, RMSEA = .10). An exam-
one's mood. The scale is also sensitive ination of the individual factor load-
to momentary changes in subject's ings indicates that all are significant
self-esteem resulting from experi-and greater than the .40 recom-
mental manipulations. Participantsmended value.
were instructed to answer fourteen We also examined a two-factor so-
lution to rule out alternative models.
items based on how they felt "at that
Therefore, they are especially sensi-

Although the two-factor state self-es-
teem model provided an acceptable
tive to social cues and positive or neg-
level of fit (CFI = .98, TO = ative
.97,contextual
IFI events. While meas-
= .98, NFI - .97, RMSEA = .09), the
ures of state self-esteem refer to a
level of fit was very close to the one- feelings of self-worth imme-
person's
factor solution. Therefore, we chose toafter some event has occurred,
diately
utilize the most parsimoniousmeasuresmodel of dispositional self-esteem
(Overall State Self-Esteem, Mean = stability refers to a person's natural
7.24, SD = 1.79, alpha = .91). tendency to have fluctuations in their
Self-esteem Level. To help control for self-esteem level. Controlling for this
alternative explanations to our find- variable allows us to rule out the idea
ings, participants' level of self-esteem that individuals with unstable self-es-
(or global self-esteem) was measured teem are simply more likely to react
using a scale developed by Rosenberg more negatively to the scenario by re-
(1986). This frequently used instru- porting lower state self-esteem.
ment for measuring a person's stable, Participants were instructed to
dispositional self-esteem has consis- complete a modified Rosenberg's
tently exhibited strong internal con- (1986) scale twice daily (e.g., 10:00
sistency and stability in past research. a.m. and 10:00 p.m., or approxi-
Respondents answered ten items mately ten to twelve hours apart) for
measuring self-esteem on a seven- a period of four days (Monday -
point scale (1 = strongly disagree, 7 Thursday, or Tuesday - Friday) dur-
= strongly agree). In addition, they ing the week immediately following
answered the questions based on how the completion of their measurement
they feel "in general." Sample items of global self-esteem and one to two
include "I wish I could have more re- weeks prior to their participation in
spect for myself and "On the whole, the laboratory session. In addition,
students were instructed to answer
I am satisfied with myself." Consistent
with past research, the ten items werethe questions based on how they felt
summed to form the measure of "at that moment" using a scale of ten
global self-esteem (Mean = 57.90, dots SDanchored by "strongly agree"
= 8.23, alpha = .85). and "strongly disagree." Kernis and
To establish the discriminant valid- his colleagues have consistently used
ity of the global and state self-esteem this approach in measuring self-es-
scales, we conducted a confirmatory teem stability. This method has been
factor analysis. The results suggestfound to be superior to other meth-
that although the items measuringods for measuring self-esteem stability
global and state self-esteem were cor- (Kernis et aL, 1989; Marsh, 1993).
related (r = .35), the items load sep- Stability of self-esteem was assessed
arately on two distinct factors. using a standard deviation measure.
The greater the standard deviation in
Self-esteem Stability. To further con-
trol for alternative explanations tothe person's self-esteem at the various
our findings, we controlled for the measure points, the greater their insta-
dispositional stability of the partici-bility in their feelings regarding self-es-
pants' self-esteem. Individuals withteem. The lower the standard devia-
unstable (or "fragile") self-esteemtion, the more stable the person's
perceive that their self-worth is alwaysself-esteem. Kernis and his colleagues
"on the line" (Kernis et aL, 1989). (1989) have used this method in past
research. In thisdifferences
sample, betweenself-esteem
the two groups
stability averaged 5.67
in terms of global(SD = (t3.97).
self-esteem = .04,
n.s.), self-esteem
Interactional Justice. Since stability
the (t = -.53,
par-
ticular scenario we used involved the n.s.), or work experience (t = -.28,
employee receiving hostile feedback n.s.). The means, standard devia-
from the supervisor (i.e., abusive su-tions, and correlation matrix for the
pervision) and included the supervi- variables in this study are presented
in Table 1.
sor stealing the idea suggested by the
employee (i.e., a possible violation of Hypothesis 1 states that individuals
perceived fairness), we attempted towho experience abusive supervision
control for the theft of the idea using
would report lower levels of state self-
two items (1 = strongly disagree; 5 esteem
= than individuals who did not
strongly agree) adapted from Moor- experience such an episode. An or-
man's (1991) scale measuring inter-dinary least square regression analysis
actional justice (Mean = 2.49, SD =indicates a strong condition effect on
one's state self-esteem (Table 2). Af-
1.38, alpha = .92): "Your supervisor
shows concern for your rights as anter controlling for one's level of self-
employee" and "Your supervisor esteem, stability of self-esteem, inter-
actional justice, gender, and work
takes steps to deal with you in a truth-
ful manner." We included these experience, the experimental condi-
items to control for the possible tionef-explained an additional 2.3% of
fects perceptions of injustice may the variance
be in overall state self-es-
having on a person's state self-esteem.teem (p < .01). Specifically, individ-
Demographics. The genderuals andin the abusive supervision con-
work experience of the subjects dition
werereported lower levels of state
controlled to rule out any potential self-esteem than individuals in the
bias in the results. Research on self- neutral supervision condition. This is
esteem and gender has found that especially interesting given that there
were no significant differences be-
males, in general, have slightly higher
levels of self-esteem than females tween the two groups regarding their
global self-esteem or self-esteem sta-
(Kling et al., 1999) . We also wanted to
control for any potential effects the bility. Therefore, the results achieved
appear to be due to perceptions of
length of work experience might
have on the participants' reaction abusive
to supervision. Hypothesis 1 is
supported.
the scenario. Individuals with greater
levels of work experience are more Hypothesis 2 states that one's gen-
der will moderate the relationship be-
likely to have experienced an episode
tween abusive supervision and one's
of abusive supervision (either person-
ally or vicariously through others)state self-esteem. To test this hypoth-
esis, we utilized ANCOVA, control-
and this may influence their reaction
to the experimental scenario. ling for global self-esteem, self-esteem
stability, work experience, and per-
RESULTS ceptions of interactional justice. Con-
sistent with Hypothesis 2, the inter-
All analyses were checked foraction vio- term is significant (F = 6.79,
lations of the assumptions of theρ nor-
< .01). Specifically, in comparison
mal error regression model. Intoaddi- male subjects, female subjects in
tion, there are no significant this sample reported lower levels of
•κ·
•Χ·
JO
^f.
^f. ^^ fî
ON CM
(Ν(Ν
Ο νο
Ο
I frv • ^ m ^ in
•Κ·
• ο ^- · *- · *- '
1
ο
ι* ι*
V
no ri írT ^Η τ-Η |> VO Tt
οο en ο Ο Ο Ο
ί
#o •Χ· ^· -3f
^ ^· ·3€· ^·
.S ^^ -Χ- -Η- ^· -5f

_d cN^oommo^H
os en es »η es ρ ρ
Ο
Έ m m oo ^t on
1 ι> es on en j r^ ^
Ι d
h ο σ' o' ^t »η λ
^H en on vo ^ j 1-H in ^3
^H S r^r^incs'csin j <=>
Ι m es g
"3
• ·
î il
3 S H t ^ ^ «
ce Ï3 S 'S #ö <u * <υ
Η
dû ë S δ χ/ 2
, ΐΐΐΐι ! ri
, I Illllll
ä
Λ Χ)
Table 2: Abusive Supervision Effe

Regression Analysis8
Model 1 Model 2
Variable Std. Beta Std. Beta
Step I
Global Self-esteem .25*** .26***
Self-esteem Stability -.18** -.18**
Interactional Justice .51*** .15
Work Experience .06 .04
Gender .08 .08
Step 2
Abusive Supervision « -.39**
(i.e., Condition)
Total R2
.42*** .44
Change in R2b - .02**
a**p<.01,***p<.001
b Change in R2 reported for addition of "Condition"
state self-esteem following no differences inan abusive

the global level or sta-
supervision episode bility (Figure I).across
of self-esteem Please
the two ex-
note that there were perimental groups. However, the
no significant
differences between males and fe- group in the abusive supervision situ-
males in their global self-esteem
ation reported lower levels of state self-
(means = 58.52 and 57.35, respec-
esteem. Therefore, our results can be
tively; t = -.98, n.s.) or self-esteem interpreted
sta- to be a result of abusive su-
bility (means = 5.12 and 6.17, respec-
pervision rather than low self-esteem
tively; t = 1.76, n.s.). Therefore, individuals
the simply reporting lower lev-
results achieved cannot be inter-
els of state self-esteem, or dispositional
preted to have come from men and
unstable self-esteem individuals react-
women simply having different levels
ing more severely to the scenario and
of global self-esteem or dispositional
thereby reporting lower levels of state
stability of self-esteem. self-esteem.
This article also adds to the litera-
DISCUSSION
ture because it examined a moderat-
The results of our study support ing influence,

the gender, on the abusive
notion that abusive supervision supervision
can - state self-esteem relation-
have a negative effect on a person's ship. To our knowledge, no other
state self-esteem. Recall that there were study has examined a moderating in-
Figure I: Overall State Self-esteem and Gender11' b
CONDITION
a Neutral = neutral interaction with supervisor; Abusive = abusive supervisio

b Gender effects after controlling for global self-esteem, self-esteem stability,
work experience, and interactional justice.
fluence on this relationship. We found

self-esteem stability, perceptions of in-
teractional justice, and work experi-
that a person's gender moderated the
ence. Therefore, even when taking a
effects of abusive supervision on self-
esteem. Specifically, female subjects'
conservative analytical approach, gen-
der appears to be an important varia-
state self-esteem was more likely to be
influenced negatively by abusive super-
ble to explore in the effects of abusive
vision than males' state self-esteem. supervision.
This is especially important consider-
A primary source of women's self-
esteem is reflected appraisals (see
ing there were no significant differ-
ences between males and females in theory of Hypothesis 2 above). That
their global level or stability of self-es-
is, women are hyper-vigilant for feed-
teem. In addition, the interactive ef- back on their work performance and
fects for gender were found over and acceptance, and tend to be more sen-
above the effects of global self-esteem, sitive to others' appraisals of them.
When women experience

ios. A critic could anargue abusive
that it lacks
realism-because
relationship at work from we are theaskingvery
the
person who they seek
subjects how theyan wouldorganiza-
react to a hy-
tionally-sanctioned appraisal
pothetical situation. However,from,Green-
their supervisor berg
- it and is,
Eskewmetaphori-
state, "the value of
cally, a chink inroletheir armor.
playing studies The
lies not in assum-
abused individual may
ing the askwould
people really herself:
do what
Why am I not respected
they say, but in in the
learning what work-
they say
place? Do I have a
theyright
would do" to (1993:my posi-
225). Specifi-
tion and status? In sum, it could be cally, this type of methodology allows
that women's self-esteem is damaged researchers to learn about basic psy-
at a greater rate because their self- chological processes. In addition, the
perceptions legitimate old stereo- scenarios reflected a work setting that
types of women as less essential mem- was appropriate for the subjects (i.e.,
bers of the workforce (Haslanger, a part-time service job). We hasten to
1996). Abusive supervision, then, is note that we found abusive supervi-
more damaging for those who seem- sion influenced subjects' state self-es-
ingly have more to lose in a supervi- teem when they were playing the role
sor-subordinate relationship when it of an employee in a hypothetical sit-
comes to self-esteem - women. uation. It is likely that the effects re-
ported here would be even stronger
Limitations when someone actually experiences
abusive supervision.
There are several limitations to
these studies that must be addressed. CONCLUSION
First, this particular study utilizes a
student population in a laboratory This study adds to the literature
setting which may weaken the gener- which has established that abusive su-
alizability of the results (Sackett andpervision results in a number of dys-
Larson, 1990). However, it has been functional consequences. Specifi-
argued that field studies are no more cally, we found that employees who
generalizable than laboratory studies experienced an episode of abusive su-
due to the narrow range of subjects pervision were lower in state self-es-
teem than those who did not experi-
in field studies (just like labs) and be-
cause of the over-reliance on conven- ence this type of mistreatment. This
ience samples (Mitchell, 1985). In ad- finding has implications both for the
dition, Anderson, Lindsay, and well-being of employees and the func-
Bushman (1999) recently demon- tioning of organizations. People with
strated a high degree of consistency high self-worth are more satisfied
between the results in laboratory and with their jobs and simply perform
field settings. Generalizability of better (Judge and Bono, 2001). In ad-
these results may not be a problem dition, diminished self-esteem in
since business students responding to women has been found to be a critical
a business scenario are also likely to factor underlying depression (Crow-
see such scenarios as appropriate and ley, 1993).
realistic. Secondly, we found support for the
A second limitation to this study idea that gender plays a role in trans-
could be the exclusive use of scenar- lating the effects of abusive supervi-
should continue
sion into discriminate consequences to examine the in-
for subordinates. Women were found terplay of abusive supervision and
to suffer greater decrements in their gender on a variety of outcome vari-
state self-esteem following an abusiveables, perhaps including promotion
supervision episode. Because workersrates and job performance. To ad-
who have high self-esteem tend to risevance practitioner knowledge it
to the challenges organizations pres-should be determined whether abu-
sive supervision has the potential to
ent and to seek out loftier goals, it fol-
lows that abusive supervisors may beharm not only female subordinates'
stifling the productivity of their sub- self-concept but also their real pro-
ordinates, especially women, andgress in organizations, that is, their
damaging the effectiveness of organ- advancement and performance eval-
izations in the process. Future studies uations.
References
Anderson, C. Α., J. J. Lindsay and B. J. Bushman. 1999. "Research in the Psyc

logical Laboratory: Truth or Triviality?" Current Directions in Psychological Scien
8: 3-9.
Ashforth, B. 1997. "Petty Tyranny in Organizations: A Preliminary Examinat
of Antecedents and Consequences." Canadian Journal of Administrative Sciences
14: 126-140.
Bjorkqvist, K., K. Osterman and M. Hjelt-Back. 1994. "Aggression Among Uni-
versity Employees." Aggressive Behavior 20: 173-184.
Brockner, J. 2002. "Making Sense of Procedural Fairness: How High Procedural
Fairness can Reduce or Heighten the Influence of Outcome Favorability."
Academy of Management Review 27: 58-76.
Lexington Books.
Cooley, C. H. 1902. Human Nature and the Social Order. New York, NY: Scribner.
Crocker, J. and C. T. Wolfe. 2001. "Contingencies of Self-worth." Psychological
Review 108: 593-623.
Crowley,J. D. 1993. Silencing the Self : Women and Depression. New York, NY: Harper
Perennial.
Dasborough, M. T. and Ν. Μ. Ashkanasy. 2003. A Qualitative Study of Cognitive
Asymmetry in Employee Affective Reactions to Leadership Behaviors." Paper
presented at the Annual Academy of Management Meetings, Seattle, Wash-
ington.
Duffy, M. K. and W. J. Ferner. 2003. "Birds of a Feather...? How Supervisor-
Subordinate Dissimilarity Moderates the Influence of Supervisor Behaviors on
Workplace Attitudes." Group and Organization Management 28: 217-248.
Fedor, D. B., W. D. Davis, J. M. Maslyn and K. Mathieson. 2001. "Performance
Improvement Efforts in Response to Negative Feedback: The Roles of Source
Power and Recipient Self-esteem." Journal of Management 27: 79-97.
Gecas, V. 1982. "The Self-concept." Annual Review of Sociology 8: 1-33.
Greenberg, J. and D. E. Eskew. 1993

Research." Journal of Management 19
Haslanger, S. 1996. "Female Realit
Chapter in Women, Knowledge, and R
Eds. A. Garry and M. Pearsall. New
Heatherton, T. F. and J. Polivy. 199
for Measuring State Self-esteem. "Jou
895-910.
Tames, W. 1890. The Pnnciples of Psychology. New York, NY: Holt.
Judge, T. A. andj. E. Bono. 2001. "Relationships of Core Self-evaluations Traits
- Self-esteem, Generalized Self-efficacy, Locus of Control, and Emotional
Stability - with Job Satisfaction and Job Performance: A Meta-analysis."/owr-
nal of Applied Psychology 86: 80-92.
Keashly, L. 1998. "Emotional Abuse in the Workplace: Conceptual and Empirical
Issues." Journal of Emotional Abuse 1: 85-99.
A Preliminary Investigation." Violence and Victim

Kernis, M. H., B. D. Grannemann and L. C. Bar
of Self-esteem as Predictors of Anger Arousal an
ality and Social Psychology 56: 1013-1023.
Kling, Κ. C, J. S. Hyde, C. J. Showers and Β. Ν
ences in Self-esteem: A Meta-analysis." Psycho
Leary, M. R., E. S. Tandor, S. K. Terdal and D.
an Interpersonal Monitor: The Sociometer Hyp
and Social Psychology 68: 518-530.
Lind, Ε. Α., L. Kray and L. Thompson. 2001. "Pr
ments: Testing Predictions from Fairness He
Behavior and Human Decision Processes 85: 189-210.
Marsh, H. W. 1993. "Self-esteem Stability and Responses to the Stability of Self
Scale." Journal of Research in Personality 27: 253-269.
Masuch, M. 1985. "Vicious Circles in Organizations." Administrative Science Quar-
terly 30: 14-33.
McAllister, D. J. and G. A. Bigley. 2002. "Work Context and the Definition of
the Self: How Organizational Care Influences Organization-based Self-es-
teem." Academy of Management Journal 45: 894-904.
McGrath, J. E. 1964. "Toward a Theory of Method' for Research on Organiza-
tions." Chapter in New Perspectives in Organizational Research. Eds. W. Cooper,
H. Leavitt and M. Shelly. New York, NY: Wiley, pp. 533-556.
Mitchell, T. R. 1985. "An Evaluation of the Validity of Correlation Research
Conducted in Organizations." Academy of Management Review 10: 192-205.
Moorman, R. H. 1991. "Relationship Between Organizational Justice and Or-
ganizational Citizenship Behaviors: Do Fairness Perceptions Influence Em-
ployee Citizenship." Journal of Applied Psychology 76: 845-855.
Neuman, J. H. and R. A. Baron. 1997. "Aggression in the Workplace." Chapter
in Antisocial Behavior in Organizations. Eds. R. A. Giacalone and J. Greenberg.
Thousand Oaks, CA: Sage Publications, pp. 37-67.
Romano, C. 1994. "Workplace Violence Takes a Deadly Turn." Management Re-
view 83: 5.
Rosenberg, M. 1986. Conceiving the Self. Malabar, FL: Robert E. K

Company.
Sackett, P. R. and J. R. Larson. 1990. "Research Strategies and Tactics in Indus-
trial and Organizational Psychology." Handbook of Industnal and Organizational
Psychology, Vol. 1 (2nd ed.). Eds. M. D. Dunnette and L. M. Hough. Palo Alto,
CA: Consulting Psychologists Press, pp. 419-489.
Schroth, H. A. and P. P. Shah. 2000. "Procedures: Do We Really Want to Know
Them? An Examination of the Effects of Procedural Justice on Self-esteem."
Journal of Applied Psychology 85: 462-471.
Schwalbe, M. L. and C. L. Staples. 1991. "Gender Differences in Sources of Self-
esteem." Social Psychology Quarterly 54: 158-168.
Tepper, B. J. 2000. "Consequences of Abusive Supervision." Academy of Manage-
ment Journal 43: 178-190.
Relationships Between Coworkers' Organiza

Fellow Employees' Attitudes." Journal of Appli
Vermunt, R., D. Van Knippenberg, B. Van Kn
"Self-esteem and Outcome Fairness: Differe
and Outcome Considerations." Journal of App
Whitley, B. E. 1983. "Sex Role Orientation an
analytical Review.11 Journal of Personality and
Wiseman, D. B. and I. P. Levin. 1996. "Compa
Conditions of Real and Hypothetical Conseq
and Human Decision Processes 66: 241-250.

Pittsburg State University Journal of Managerial Issues

Uploaded by

Copyright:

Available Formats

Pittsburg State University Journal of Managerial Issues

Uploaded by

Document Information

Original Description:

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

Pittsburg State University Journal of Managerial Issues

Uploaded by

Copyright:

Available Formats

Subordinate Self-esteem and Abusive Supervision

Author(s): James P. Burton and Jenny M. Hoobler

Subordinate Self-esteem and Abusive Supervision*

While physical acts of workplace vi-healthy workforce. Patterns of nega-

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

THEORY AND BACKGROUND Abusive Supervision and

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

positive feedback, counters

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

may be more likely ratoryto session scheduled

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

participants read the first part ofHethe

signed to one of two conditions. In

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

Participants moment," having just read the sce-

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

Therefore, they are especially sensi-

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

.S ^^ -Χ- -Η- ^· -5f

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

Table 2: Abusive Supervision Effe

Change in R2b - .02**

state self-esteem following no differences inan abusive

The results of our study support ing influence,

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

Figure I: Overall State Self-esteem and Gender11' b

a Neutral = neutral interaction with supervisor; Abusive = abusive supervisio

fluence on this relationship. We found

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

When women experience

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

Anderson, C. Α., J. J. Lindsay and B. J. Bushman. 1999. "Research in the Psyc

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

Greenberg, J. and D. E. Eskew. 1993

A Preliminary Investigation." Violence and Victim

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

Rosenberg, M. 1986. Conceiving the Self. Malabar, FL: Robert E. K

Relationships Between Coworkers' Organiza

JOURNAL OF MANAGERIAL ISSUES Vol. XVIII Number 3 Fall 2006

You might also like