The Salieri Syndrome: Consequences of Envy in Groups

Abstract

A model of the impact of envy in groups is proposed and tested in a longitudinal study of 143 groups. Envy was directly and negatively related to group performance. Moreover, envy indirectly influenced group performance, absenteeism, and group satisfaction by increasing social loafing and reducing both group potency and cohesion. This study provides an initial step in identifying the processes through which envy impacts group effectiveness. Implications are discussed and future research directions are identified.

Full text

http://sgr.sagepub.com/
Small Group Research
http://sgr.sagepub.com/content/31/1/3
The online version of this article can be found at:
DOI: 10.1177/104649640003100101
2000 31: 3Small Group Research
Michelle K. Duffy and Jason D. Shaw
The Salieri Syndrome: Consequences of Envy in Groups
Published by:
http://www.sagepublications.com
can be found at:Small Group ResearchAdditional services and information for
http://sgr.sagepub.com/cgi/alertsEmail Alerts:
http://sgr.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://sgr.sagepub.com/content/31/1/3.refs.htmlCitations:
What is This?
- Feb 1, 2000Version of Record >>
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

SMALL GROUPRESEARCH / February 2000Duffy,Shaw / THE SALIERI SYNDROME
THE SALIERI SYNDROME
Consequences of Envy in Groups
MICHELLE K. DUFFY
JASON D. SHAW
University of Kentucky
A model of the impact of envy in groupsisproposedandtestedinalongitudinalstudyof143
groups.Envywasdirectly andnegativelyrelatedtogroup performance.Moreover,envyindi
-
rectly influenced group performance, absenteeism, and group satisfaction by increasing
social loafing and reducingboth group potencyand cohesion. This study provides aninitial
step in identifying the processes through which envy impacts group effectiveness. Implica
-
tions are discussed and future research directions are identified.
Whodarestosaythattheeverproud Saliericould stoopto envy, like
a loathsome snake, trampled upon by men, yet still alive and impo-
tently gnawing sand and dust?...Butnow—myself I say it—yes
now I do know envy—Yes, Salieri envies! O Mozart, O Mozart!
—Pushkin (1832/1964)
Until recently, the experience of negative emotion at work
received relatively little attention from social science researchers
(George, 1990, 1992a). However, interest in the study of negative
emotion and its consequences is growing as researchers recognize
thecumulativelynegativeeffects these emotionshaveonimportant
organizational outcomes such as performance, employee with
-
drawal, sabotage, and turnover (e.g., Ashforth & Lee, 1990). One
potentially important, yet virtually ignored, emotion deserving
attention in organizational research is the experience of envy
(Bedeian,1995;Vecchio, 1995).Althoughscarcelyresearched, the
experience of envy has a long, colorful history, including the infa
-
mous story of Mozart and his contemporary, Salieri.
3
SMALL GROUP RESEARCH, Vol. 31 No. 1, February 2000 3-23
© 2000 Sage Publications, Inc.
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

ENVY DEFINED
Often confused with jealousy, envy occurs when the perception
exists that a “person lacks another’s superior quality, achievement,
orpossession and eitherdesiresit or wishesthatthe other lacked it”
(Parrott & Smith, 1993, p. 906). According to these authors, the
experienceofenvy isbestconceptualized asaconstellation ofvari
-
ous distinguishable affective elements that typically occur during
episodes of envy. Such affective reactions may include, but are not
limited to, inferiority, longing, resentment of one’s circumstances,
shame,depression,helplessness,insecurity,frustration,andill-will
toward the envied person.
Asthe previous constitutivedefinitionsuggests,social compari
-
sonwithone’speersisoftenthetriggerforenvy.Giventhatmuchof
one’s self-worth is derived from social comparison (see Tesser &
Campbell, 1990, for an example), it is not surprising that when
one’s accomplishments, talents, and possessions are perceived to
comparepoorlywith thoseofanotherindividual,privateandpublic
self-esteem may suffer (Parrott, 1991). In addition to envy arising
from decreases in esteem and public stature, negative social com-
parison is theorized to lead to envy by “heightening one’s aware-
ness” that “one’s own deprivation” and suffering are not shared by
all (Parrott, 1991, p. 7).
Despite the fact that little systematic research examining the
antecedents and consequences of envy in the workplace is con
-
ducted (see Vecchio, 1995, for an exception), it is easy to imagine
how the nature of organizational life gives rise to frequent social
comparisons that may give rise to envy. For example, competition
for and allocation of scarce organizational rewards in the form of
merit raises, office space, promotions, grants, valued assignments,
and promotions are all potential catalysts for social comparison
among colleagues. Merely noticing the superior work achieve
-
ments of another may also result in envious social comparisons
betweenoneselfandacolleague(Tesser&Campbell,1990).Asthe
Cinderella myth suggests, a person who is simply striving to meet
personal standards may arouse the resentment of others merely for
that fact alone (Ulanov & Ulanov, 1983). Finally, as Bedeian
4 SMALL GROUP RESEARCH / February 2000
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

(1995) notes, any situation in which one employee obtains an
advantage at the expense of other employees may provoke
coworker envy.
CONSEQUENCES OF ENVY
Feelings of envy often lead to a variety of affective and behav
-
ioral reactions that generally are believed to be aimed at repairing
damaged self-esteem and social status (Salovey & Rothman,
1991). Results from studies in social psychology suggest that such
damage control–based reactions often occur at the expense of the
individual and/or the rival. Responses to envy include depressed
mood and anxiety, avoidanceof the comparisonperson (Salovey &
Rodin, 1986), overt hostility (Parrott & Smith, 1993), degrading
comparison individuals and their accomplishments (Silver &
Sabini, 1982), and attempting to prevent the rival’s successful per-
formance(Tesser&Smith,1980).WhiteandMullen(1989)identi-
fiedninemajorcopingstrategiesfordealingwithjealousyandenvy
(family, friend, and romantic), which Vecchio (1995) then synthe-
sized with the organizational setting. According to Vecchio’s
work-adaptedmodel,potential reactions to envymayincludesabo-
taging the rival’s work, back-stabbing a competitor, harassment or
ostracismoftherival,pretendingtobedisinterestedintherival,and
bolstering one’s own self-image.
Unfortunately, although envy-related reactions receive some
empiricalattentioninthesocialpsychologyliterature,weareaware
of only one empirical study to date (Vecchio, 1995) that examined
the effect of envy on outcomes typically considered relevant in
work contexts. Results of this study indicated that envy of one’s
colleagues was significantly related to propensity to quit, job dis
-
satisfaction, and supervisor dissatisfaction. In sum, although envy
is a pervasive and commonplace emotion with important implica
-
tions for understanding organizational behavior, researchers only
“reluctantlyacknowledgedtheinhibitinganddestructiveaspectsof
this emotion in the workplace” (Bedeian, 1995, p. 50).
Duffy, Shaw / THE SALIERI SYNDROME 5
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

Giventhe seriousconsequencesresultingfromenvy-ladeninter
-
personal relationships, it is imperative that employees who are
experiencing negative emotions such as envy “be the object of
study inorder to moreaccurately determine whatprotagonists, tar
-
gets,andrivalsactuallydo,feel,andthink”(Vecchio,1995,p.216).
Considering the increase in employee violence and workplace
sabotage, as well as the increasing costs of deviant workplace
behavior (e.g., Robinson & Bennett, 1995), calls for further
research are clearly warranted. This article is intended to begin to
fill this gap by exploring the consequences of the experience of
envy in groups.
ENVY IN THE GROUP CONTEXT
Vecchio (1995) suggests that the recently popular approach to
increasing organizational commitment through the use of self-
managed work teams may result in diminished feelings of work-
place jealousy and envy. Briefly, Vecchio theorizes that shared
responsibility for task completion and mutual work dependence in
groups may serve to attenuate the competitiveness found in tradi-
tional workplace settings. Although this may be true, it should be
recognized that work groups may also be an ideal setting for inci-
dences of envy in the workplace. That is, the occurrence of envy in
groups may be especially likely because group members work
closely together. The result is that group members interact often
and come to know each other very well. As Tesser and Campbell
(1990) suggest, the closer an individual is to someone, the more
likely that envious comparison processes will occur. Envy in
groups may be especially detrimental given the proclivity of envi
-
ous individuals to engage in deviant behaviors (e.g., socialloafing,
sabotage), which may cause harm not only to themselves, but to
other group members. The ultimate result may be a diminishment
of group performance and other positive group outcomes. These
factors,combined withthe fact thatorganizationsare continuingto
emphasizeteam-orgroup-basedwork,suggestthatresearchexam
-
ining envy in work groups is a critical area for study.
6 SMALL GROUP RESEARCH / February 2000
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

THE PRESENT STUDY
Figure 1 depicts the conceptual framework underlying this
study. Envy in the work group is proposed to influence group out
-
comes (e.g., group performance, absenteeism, and group satisfac
-
tion) both directly and indirectly. Drawing on theory from social
psychology, wepropose thathigher levelsof envy within thegroup
result in lower levels of group performance and group satisfaction
and higher absenteeism rates. The experience of envy has been
described as a consuming emotion in which one’s weaknesses or
flaws feel physically apparent. Low self-esteem, anxiety, hostility,
depression,or decreasedself-efficacymay result. Tothe extent that
envy interferes withthe abilityto perform,increases inenvy within
the group should be associated with decreases in group perform-
ance. Inaddition, becausewithdrawalis often acoping strategy for
dealing with envy, it is predicted that higher levels of envy would
result in increased absenteeism. Finally, if group members are
increasingly resentful of one another and of each other’s accom-
plishments, group dissatisfaction should be another consequence.
In addition to these direct effects, we propose that envy indi-
rectly influencesgroup outcomesthrough social loafing, cohesion,
andpotency.First,drawingonsocialpsychologyliterature,wepro-
pose that increased levels of envy will be associated with higher
levels of social loafing which, in turn, will be detrimental to group
functioning (Latané, Williams, & Harkin, 1979). As previously
noted, responses to envy are believed to include avoidance of the
comparison person(s), hostility, and attempts to prevent a rival’s
successful performance (Parrott & Smith, 1993). This occurs even
at the expense of one’s own success. Social loafing behavior repre
-
sents an ideal method of simultaneously sabotaging a rival’s per
-
formance (by refusing to participate in the group effort) and mani
-
festinghostility in apassive-aggressive manner(e.g., notfollowing
through on group-related commitments) (George, 1992b; Latané
et al., 1979). These behaviors then would be expected to diminish
group effectiveness.
Envy is also proposed to be associated with decreased levels of
group cohesiveness and potency, which are then hypothesized to
Duffy, Shaw / THE SALIERI SYNDROME 7
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

influence group outcomes. Cohesiveness includes elements of
interpersonal attraction and group commitment and is consistently
positivelyrelatedto groupeffectiveness(e.g.,Evans& Dion,1991;
Gully, Devine, & Whitney, 1995; Mullen& Copper, 1994;Reizen-
stein & Burke, 1996). Groups that direct more energy toward the
desired outcomes through increased commitment, rather than
toward internal conflict and conflict management, should perform
moreeffectively(Reizenstein& Burke,1996;Wolfe&Box,1988).
Envy is associated with higher levels of hostility, anxiety, and
back-stabbing (Parrott & Smith, 1993; White & Mullen, 1989).
Such feelings could be expected to decrease the “we” feeling of
group members captured by group cohesion.
Furthermore, the characteristic back-stabbing attitudes and
behaviorsof envious group members should also diminish the per
-
formance expectations or potency (Shea & Guzzo, 1987) of the
group. Potencyis the belief that the group can beeffective (Guzzo,
Yost, Campbell, & Shea, 1993) and can be thought of as a type of
self-efficacy or high expectancy for the group (Campion, Medsker, &
Higgs,1993). Hackman(1987) proposed thatgroup potency facili
-
tates commitmentto the group and greater effort from groupmem
-
bers, ideas that are completely at odds with envious emotion.
8 SMALL GROUP RESEARCH / February 2000
Figure 1: Hypothetical Model
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

Obviously, if envy results in higher levels of social loafing, group
potency would be diminished and group effectiveness would be
compromised. Moreover, as envy redirects energy away from
group activities, the belief of the group as a whole that it can suc
-
ceedmaybe diminished,reducing group effectiveness.Itshould be
explicitly noted that the focus and context of our study is on the
emotion or stateof envyelicited bythe currentmembers ofan indi
-
vidual’s group. The study of envy, although sparse, also includes
the examination oftrait or dispositional enviousness(e.g., Smith&
Turner, 1996), which is beyond the scope of this study.
METHOD
SAMPLE
Participantsin thestudy were566 upper-division undergraduate
students enrolledin group-basedclasses ata largesouthern univer-
sity. The sample included 143 groups ranging in size from three to
seven members. The sample was 61% male with a mean age of 22
years. Three self-report questionnaires were administered at three
time periods over a 16-week term. Participants were guaranteed
confidentiality and were assured that participation was voluntary.
Each participant in the study signed a waiver that allowed the
research team to collect performance information from class
instructors atthe end ofthe term. The data collectionswere tempo
-
rallyseparatedbyaperiodof8weeks,inparttoreducetheresponse
bias associated with common method and fatigue from long ques
-
tionnaire administration (Ganster, Fusilier, & Mayes, 1986). The
first data collection occurred during the first week of class and
before participants were assigned into groups. In this collection,
background information and ability (e.g., grade point averages)
measures were collected.The secondwave ofdata was collectedin
the week following midterm, about 8 weeks after the first admini
-
stration.Thiscollectionincludedtheenvy,socialloafing,cohesive
-
ness, and potency measures. Explicit oral instructions at Time 2
Duffy, Shaw / THE SALIERI SYNDROME 9
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

cued participants to respond to questionnaire items with regard to
their assigned work group in the focal class only. Moreover, the
writteninstructionsets foreachrelevantsectionoftheTime2ques
-
tionnaire prompted participants to respond to the group items with
reference tothe theircurrent group inthat class.Group satisfaction
and absenteeism measures were collected in the final data collec
-
tion during the last week of class or approximately 16 weeks after
the first questionnaire administration. Following the term, group
performance and group size for each group was collected from
course instructors.
GROUP CONTEXT
Participantswereenrolled incoursestaught inaformat inwhich
a large part of an individual’s course grade was based on the per-
formance of their assigned group. All of the course instructors
required groups to work on projects during and outside of class
time. Participants reported working with their group outside of
class for an average of 2.27 hours per week during the term, or
roughly 36 hours outside of class in addition to in-class projects.
Thus, although the use of undergraduate students to simulate work
environments hasbeen the subject of muchdebate, theparticipants
had substantial interactions with group members over a 4-month
period in a simulated work-group environment. These simulated
conditions are similar to the context in a recent study by Wagner
(1995).
MEASURES
All measures were operationalized at the group level. Thus,
exceptwhere noted,the mean levelfor thegroup formsthe mea-
sure. Also, except where noted, all items were in a Likert-type for
-
mat with seven response options. Coefficient alpha reliabilities for
the scales are reported on the main diagonal in Table 1.
Independent variables. Envy was measured with five items
adapted from a scale created by Vecchio (1995). The items were
10 SMALL GROUP RESEARCH / February 2000
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

adaptedtoreflectfeelingsofenvywithrespecttoone’sgroupmem-
bers in this study. These items were designed to assess feelings of
resentment and inferiority relative to group members. A sample
item is, “Most of my team members haveit betterthan Ido.” Social
loafing was measured with three items adapted from a scale by
George(1992b).The measure tapsself-reportsocialloafingbehav-
ior. Theitems examined theextentto which anindividualtended to
do less than his or her share of work when other group members
were available to do the work. A sample item is, “Sometimes I let
myteam membersdothe work thatI should do.” Cohesivenesswas
measuredwithafour-itemscaledevelopedbySeers(1989).Asam
-
ple item is, “This group has a strong sense of togetherness.” Group
potencywasmeasuredusinganeight-itemscaledevelopedbyShea
and Guzzo (1987). A sample item is, “This group believes it can
produce extraordinary work.”
Control variables. We included two control variables that may
be related to the independent or dependent variables: group size
and ability. Groups should be large enough to handle group tasks,
butsmallenough tocontrol coordinationneeds (e.g.,Campion et al.,
1993; Gladstein, 1984), and consequently, group size may be
related to performance and group process. Group size was col
-
lectedfromtheinstructor ofeachcourse. In addition,we controlled
Duffy, Shaw / THE SALIERI SYNDROME 11
TABLE 1: Descriptive Statistics and Correlations Among All Study Variables
MSD 123 456789
1. Group
performance 0.00 1.00 #
2. Absenteeism 3.62 9.83 –.05 #
3. Group
satisfaction 5.33 1.35 .26** .25**(.84)
4. Envy 2.92 0.71 –.28** .19* –.24**(.75)
5. Social loafing 4.43 1.03 –.20* .10 –.33** .40**(.81)
6. Cohesiveness 4.97 0.84 .20* –.25** .59**–.30**–.46**(.83)
7. Potency 5.27 0.69 .35**–.12 .54**–.33**–.26**.58**(.90)
8. Ability 3.27 0.82 .04 –.10 .01 –.12 –.06 .06 .07 #
9. Group size 4.52 1.12 .08 .12 .23** .01 .02 .05 .22** .06 #
NOTE:Pairwisedeletionprocedureusedtogeneratethecorrelationmatrix.Ns=130-143.
*p < .05. **p < .01. Two-tailed tests.
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

for the average ability level of group members utilizing a com
-
monly used proxy for ability, grade point average (Wagner, 1995).
Gross ability levels are used as a control because they may be
related to performance levels in groups (Saavedra & Kwun, 1993).
Participants reported their cumulative grade point average during
the first data collection. The mean level for the group forms the
measure. Although not the central focus of the study, inclusion of
these variables makes the results more generalizable and reduces
the potential for alternative explanations of the findings.
Groupperformance.Performanceor effectivenesscanbeopera
-
tionalized or measured on several dimensions (Campion et al.,
1993). We measured the performance of the group along three
dimensions: overall group performance, group satisfaction, and
absenteeism.Overallgroup performance representsthepart ofper-
formance in theclass solely attributableto the group. This variable
was not an aggregationof individual performance, but rather a true
group-level measure. This variable was collected from course
instructors following the term. The variable was standardized by
class for comparability across the sample. Group satisfaction was
measured with a three-item scale adapted from a scale developed
by Cammann, Fichman, Jenkins, and Klesh (1983). A sampleitem
is,“Allinall,Iamsatisfiedwithmyteam.”Absenteeismwasopera-
tionalized as the number of times participants were absent from
class over the course of the term and was reported by the partici
-
pants.Thismeasurewascollectedduringthefinaldatacollection.
ANALYSIS STRATEGY
Correlationalanalyseswereusedtoinvestigatethehypothesized
bivariate relationships. Path analysis (Cohen & Cohen, 1983) was
used to obtain path estimates for the hypothesized model. A full
path analysis for each of the three group performance outcomes
was conducted following procedures described by Cohen and
Cohen (1983). Group size and ability measures were used as con
-
trol variablesin allsteps ofthe pathanalysis. To estimatemodel fit,
the Q statistic (Pedhazur, 1982) was used.
12 SMALL GROUP RESEARCH / February 2000
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

RESULTS
Descriptive statistics and correlations among all study variables
are shown in Table 1. As expected, envy was related to each of the
three group-performance measures in the zero-order correlations.
Envy was negatively related to group performance (r = –.28, p <
.01) andgroup satisfaction(r =–.24, p <.01) and positively related
to absenteeism (r = .19, p < .05). Furthermore, the relationship
between envy and each of the intermediate variables in the model
was also significant and in the predicted direction. Envy was posi
-
tively related to social loafing (r = .40, p < .01), and negatively
relatedto cohesiveness(r =–.30,p <.01) andpotency(r =–.33,p <
.01). Moreinteresting, however,are the pathanalyses that allow an
assessment of these relationships in context.
Theresults ofthepath analysesregressionarereported inTable 2.
In each of the equations, the relationships of interest are examined
after partialling the effects of group size and ability. The results of
the three path analyses are also depicted graphically in Figures 2
through 4. For clarity, control variables are not included in the fig-
ures. In each figure, solid lines denote the significant paths.
Overall, the control variables were not strong predictors. Group
size was only related to group satisfaction (β = .16, p < .05) in the
fullequation,whereastheabilitymeasurewasnot a significantpre-
dictor in any equation. These results mirror the bivariate relation
-
ships reported in Table 1.
The path analyses indicated that envy was positively associated
with social loafing in the group (β = .36, p < .01) and negatively
relatedtopotency(β=–.34,p<.01)andcohesiveness(β=–.18,p<
.05). Interestingly, and contrary to expectations, envy was not
directly related to the group effectiveness measures, although the
relationships with group performance (β = –.16) and absenteeism
(β = .17) approached significance.
Social loafing was found to be associated with decreased
potency (β= –.18, p < .05) and cohesiveness (β = –.40, p < .01), as
expected. Although social loafing was negatively related to both
group performanceand satisfactionat the bivariate level, therewas
nodirect effectofsocial loafingon these outcomeswhen enteredin
Duffy, Shaw / THE SALIERI SYNDROME 13
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

the regression equation with othervariables.Thus, the relationship
between social-loafing and group performance measures is medi
-
ated by potency and/or cohesiveness.
The path analyses also indicated that cohesiveness was nega
-
tively related to absenteeism (β = –.29, p < .01) and positively
related to group satisfaction (β= .37, p < .01). However, there was
14 SMALL GROUP RESEARCH / February 2000
TABLE 2: Path Analysis Results
Intermediate Dependent Variables Group Performance Variables
Social Group Group
Loafing Potency Cohesiveness Performance Absenteeism Satisfaction
Control
variables
Ability .02 .01 .04 –.02 –.08 –.03
Group size .00 .15 .05 –.00 .13 .16*
Independent
variables
Envy .36** –.34** –.18* –.16 .17 .03
Social loafing –.18* –.40** –.12 –.07 –.09
Cohesiveness –.08 –.29** .37**
Potency .27* .08 .28**
Total R
2
.131** .227** .231** .147** .109* .429**
NOTE: Standardized regression coefficients are reported for all variables. N = 129.
*p < .05, two-tailed. **p < .01, two-tailed.
Figure 2: Path Model With Group Performance as the Dependent Variable
NOTE: Solid lines denote significant paths.
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

no significant relationship between cohesiveness and group per
-
formance(β=–.08, ns).Interestingly, potency was foundto besig
-
nificantlyrelated togroupperformance (β= .27,p< .05)andgroup
satisfaction(β=.28,p<.01),butnottoabsenteeism(β=.08,ns).
Lastly, as shown in Figures 2 through 4, the following indirect
pathswere foundto besignificant:(a) envy andgroup performance
through potency as well as through social loafing; (b) envy and
group satisfaction through potency, cohesiveness, and social loaf
-
ing; (c) envy and absenteeism through cohesiveness and social
Duffy, Shaw / THE SALIERI SYNDROME 15
Figure 3: Path Model With Absenteeism as the Dependent Variable
NOTE: Solid lines denote significant paths.
Figure 4: Path Model With Group Satisfaction as the Dependent Variable
NOTE: Solid lines denote significant paths.
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

loafing;(d) socialloafingand groupperformancethrough potency;
(e) social loafing and group satisfaction through cohesiveness and
potency; and (f) social loafing and absenteeism through
cohesiveness.
The goodnessof fit foreach modelwas tested using theQ statis
-
tic (Pedhazur, 1982). The Q statistic is the “ratio of explained vari
-
ance to the variance to be explained” (Pedhazur, 1982, p. 619).
More formally,
QR
M
=−
−
1
1
In theformula, R
m
2
is equalto (1– (1– R
1
2
)(1 –R
2
2
)...(1–R
p
2
)),
whereR
i
2
isthe“ordinarysquaredmultiplicationcorrelationcoeffi
-
cient of the i
th
equation in a fully recursive system” (Pedhazur,
1982, p.619). R
m
2
can beinterpreted asthe “ratioof explained vari-
anceto thevariancetobe explained” (Pedhazur, 1982,p. 619).Mis
definedasthesquaredmultiplecorrelationcoefficientforareduced
modelinwhichnonsignificantpathsaredeleted.TheQstatisticcan
vary from zero to one, with values approaching one indicating
excellentmodelfit(Pedhazur,1982).Qisnotbiasedbysamplesize
asaremeasuresoffitsuchaschi-square,andthusrepresentsamore
precisewaytoassessgoodnessoffit.Inthepresentstudy, the Q sta-
tistic for group performance and absenteeism by group were .79
and .85, respectively, indicating good model fit. Q for group satis
-
faction was .98, indicating excellent model fit.
DISCUSSION
The results of this study shed new light on how groups function
and the processes by which envy diminishes the overall group
effectiveness. Given that the cost of deviant behavior in the work
place is prohibitive (Vecchio, 1995), and because social-
psychological studies show envy to berelated toa plethoraof dam
-
aging individual behaviors and attitudes (e.g., Parrott & Smith,
1993), it is important that more be known about the influence of
16 SMALL GROUP RESEARCH / February 2000
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

envy in group dynamics. This study was aimed at investigating
whetherenvydiminishes groupeffectiveness.Moreover, thisstudy
represents a first attempt at peeking into the “black box” and
exploringhowenvyinfluencesgroupeffectiveness.Weexamineda
path model of envy’s influence using longitudinal data on 129
groups. Based on previous research at the individual level, it was
hypothesized that the level of envy in the group would increase
socialloafinganddiminishgroupcohesionandpotencythatwould,
in turn, diminish group effectiveness. The results generally sup
-
ported these expectations.
The strong relationship between envy and social loafing is con
-
sistentwith existing theory. Previous research insocial psychology
shows that envious individuals often pretend to be disinterested in
their rivals(Vecchio, 1995; White &Mullen, 1989). It appears that
thisdisinterestmay be translatedinto higher levelsof social loafing
in groups. The result is damage to the processes through which
groupsfunction effectively, which consequentlylowersthe levelof
group effectiveness. Interestingly, social loafing was not directly
relatedto anyof theoutcome measures;theeffectsof socialloafing
on performance, absenteeism, and group satisfaction were com-
pletely mediated by the group-process measures of cohesiveness
and potency.
Envy was also directly related to group cohesiveness and
potency. Cohesiveness includes both elements of interpersonal
attraction and group commitment(Gully etal., 1995),factorstheo
-
retically at odds with the back-stabbing, harassing nature of envy.
The empirical results reported here support this theoretically pre
-
dicted negative relationship. In addition, cohesiveness mediated
therelationshipbetween envy and absenteeismand envy andgroup
satisfaction. However, potency mediated the relationship between
envy and group performance. Although empirical research with
respect to potency is rather scarce, these results confirm specula
-
tionswithrespect to potency. Potencyis theoretically proposedas a
performancedimensionoratypeofefficacyorhigh-expectancyfor
the group(Campion et al., 1993), and these resultstend to bearthis
out. Higher levels of potency were strongly related to group
Duffy, Shaw / THE SALIERI SYNDROME 17
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

performance, but not to the other outcomes. Findings that show
potency to be related to performance, and cohesiveness to be
relatedtogroup-memberattitudes(e.g.,satisfaction)andbehaviors
(e.g., absenteeism) support much of the current theory on group
functioning (e.g., Campion et al., 1993).
More importantly, these results show the encompassing detri
-
mental influences of envy in groups. Envy is associated with effort
reduction inthe group, but thesabotaging nature ofenvy also dam
-
ages intragroup relations, lowering cohesion, which increases
absenteeism and results in lower levels of satisfaction. Absentee
-
ism, in particular, has a rich history in the organizational sciences
(e.g., Beehr & Gupta, 1978) and is among a set of withdrawal
behaviors that can be considered critical facets of group effective
-
ness(Blau, 1995).The finding that envyresults inincreased absen-
teeism in groups is noteworthy. Moreover, envy whittles away the
members’beliefs that they canperform effectively as agroup, low-
ering group potency, which results in lower levels of group
performance.
This study should be viewed in light of potential weaknesses.
The study utilized a student sample, a subject of much debate.
Counteracting this limitation is the fact that the group context in
this sample is similar in some ways to group contexts found in real
organizational contexts. The groups worked together over a 16-
week period for an average of 2.27 hours per week outside of the
classroom, in addition to in-class assignments. Also, the partici
-
pants were upper-division undergraduates, preparing to enter the
full-timeworkforceinthe nearfuture, andmay beconsidered simi
-
lar to a young professional workforce. Another strength of the
study is the fact that the data were collected over time and that one
of the key dependent variables, group performance, was collected
from a separate source. Absenteeism was collected through self-
reports, calling into question the validity of the measure. Partici
-
pantsin thestudywere assuredconfidentialitythroughout theproj
-
ect, and members of the research team reminded participants at
each stage of the project that instructors would not have access to
individual information. Specific instructions with regard to
18 SMALL GROUP RESEARCH / February 2000
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

confidentiality and absenteeism were given during the final data
collection. These steps should have reduced participant fear con
-
cerning reporting their absence. To the extent that all participants
underreported their absence behavior, the results would not be
biased except in the absolute magnitude of absenteeism. Predictor
variables were collected 2 months before the outcome measures.
Among the many benefits of examining dynamics over time is the
reduction in common method variance in the study (Ganster et al.,
1986). Despite these counteractions, it isobvious that similar stud
-
ies in other group contexts are needed.
The results of this study suggest several opportunities for future
research, including investigations examining the extent to which
the relationships found in this study hold over time and across dif
-
ferent contexts. The identification of alternative mediators of the
relationship between envy and group outcomes would also be an
interesting avenue for future research. Although this study exam-
ined the dynamics of envy, group process variables, and outcomes,
anotherpotentiallyfruitfulareaforfutureresearchistheetiologyof
envy. Such studies could include investigations of how and why
envious persons choose their envied comparison persons. In a
relatedvein,futureresearchcouldexploreaspects of the workplace
or a managerial style that foster envy among coworkers.
Future investigations could also focus on personality and indi-
vidual difference variables as determinants of envy, and the differ
-
ences between the emotional state of envy and dispositional envi
-
ousness. In this study, we focused on the emotion of envy, and in
particular, feelings of envy with regard to the group at hand. How
-
ever,afewstudies (e.g.,Smith& Turner,1996)do focus onthetrait
of enviousness, a stable disposition to feel envious. Such a distinc
-
tioncallsto mindthe burgeoningliterature onpositiveand negative
affectivity that is seen theoretically, and demonstrated empirically,
as both an emotion (e.g., Isen & Baron, 1991) and a dispositional
trait (e.g., Shaw, Duffy, Jenkins, & Gupta, 1999). To date, little
researchconcernsthepersonalitycomponentofenvyorthoroughly
describes a nomological network of correlates of dispositional
envy. Are some individuals more prone to envy? Do individual
Duffy, Shaw / THE SALIERI SYNDROME 19
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

differences determine whether a person will experience the emo
-
tion of envy when faced with a negative social comparison?
Although we were unable to address the dispositional component
of envy in this research, we did compare, post hoc, the profile of
participants who reported varying levels of envy. These results
showed that the relationships between envy and age, gender, class
standing (e.g., junior, senior), previous experiences in teams, and
general self-esteem were not statistically significant, but envy was
negativelyrelated togradepoint average(r= –.14,p <.01),interest
inthesubjectmatterintheclass(r=–.14,p<.01),self-efficacy(r=
–.20,p <.01) andwashigheramong thosewithan external locus of
control (r = .24, p < . 01). Althoughpreliminary and possibly more
informative with regard to the emotion of envy in this particular
context,theseresultsmaysuggestthatthestate ofenvymaybetrig-
gered both by a lack of confidence with regard to particular tasks
and a perceivedlack ofcontrol over relevantevents. Theseprelimi-
naryandexploratoryfindingsmayprovideadirectiontowardacer-
tainlyinterestingandapotentiallyfruitfulareaforfutureresearch.
1
Moreover, research examining vertical comparisons in organi-
zationsmayoffersomepotentiallyinterestinginsightsintothecon-
sequences of envy in groups and organizations. A final suggestion
for future research concerns a test of Vecchio’s (1995) proposition
that the use of work groups in organizations has the potential to
reduce feelings of envy among coworkers. Vecchio argues that
shared responsibility and mutual work dependence in groups may
attenuate workplace competition, and in turn, reduce envy. How
-
ever, social psychologysuggests that intense interactionswith oth
-
ers increases envy. A study that compares feelings of envy in the
traditional work design withenvy in a group work design would be
a highly interesting test of these competing predictions.
NOTE
1. We thank an anonymous reviewer for highlighting many of these ideas.
20 SMALL GROUP RESEARCH / February 2000
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

REFERENCES
Ashforth, B. E., & Lee, R. T. (1990). Defensive behavior in organizations: A preliminary
model. Human Relations, 43, 621-648.
Bedeian, A. G. (1995). Workplace envy. Organizational Dynamics, 23, 49-56.
Beehr, T., & Gupta, N. (1978). A note on the structure of employee withdrawal. Organiza
-
tional Behavior and Human Performance, 21, 73-79.
Blau, G. (1995). Influence of group lateness on individual lateness: A cross-level examina
-
tion. Academy of Management Journal, 38, 1483-1496.
Cammann,C.,Fichman,M.,Jenkins,G.D.,Jr.,&Klesh,J.R.(1983).Assessingtheattitudes
and perceptions of organizational members. In S. E. Seashore, E. E. Lawler, III, P. H.
Mirvis, & C. Cammann (Eds.), Assessing organizational change: A guide to methods,
measures, and practices (pp. 71-138). New York: John Wiley.
Campion, M. A., Medsker, G. A., & Higgs, C. A. (1993). Relations between work group
characteristics and effectiveness: Implications for designing effective workgroups.Per
-
sonnel Psychology, 46, 823-850.
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the
behavioral sciences. Hillsdale, NJ: Erlbaum.
Evans, C. R., & Dion, K. L. (1991). Group cohesion and performance: A meta-analysis.
Small Group Research, 22, 175-186.
Ganster, D. C., Fusilier, M. R., & Mayes, B. T. (1986). Role of social support in the experi-
ence of stress at work. Journal of Applied Psychology, 71, 102-110.
George, J. M. (1990). Personality, affect, and behavior in groups. Journal of Applied Psy-
chology, 75, 107-116.
George, J. M. (1992a). The role of personality in organizational life: Issues and evidence.
Journal of Management, 18, 185-213.
George,J.M.(1992b).Extrinsicandintrinsicoriginsofperceivedsocialloafinginorganiza-
tions. Academy of Management Journal, 35, 191-202.
Gladstein,D.L.(1984).Groups incontext:Amodeloftaskgroupeffectiveness.Administra
-
tive Science Quarterly, 29, 499-517.
Gully, S. M., Devine, D. J., & Whitney, D. J. (1995). A meta-analysis of cohesion and per
-
formance: Effects of levelof analysis and task interdependence. Small Group Research,
26, 497-520.
Guzzo,R.A.,Yost,P.R., Campbell,R.J.,&Shea, G.P.(1993). Potencyingroups: Articulat
-
ing a construct. British Journal of Social Psychology, 32, 87-106.
Hackman,J.R.(1987).The designofworkteams.InJ.W.Lorsch(Ed.),Handbook oforgan
-
izational behavior (pp. 315-342). Englewood Cliffs, NJ: Prentice-Hall.
Isen, A., & Baron, R. (1991). Positiveaffect as afactor in organizational behavior. Research
in Organizational Behavior, 13, 1-53.
Latané, B., Williams, K.,& Harkin, S. (1979). Many hands make light the work: The causes
and consequences of social loafing. Journal of Personality and Social Psychology, 37,
822-832.
Mullen, B., & Copper, C. (1994). The relation between group cohesiveness and perform
-
ance: An integration. Psychological Bulletin, 115, 210-227.
Parrott,W. G. (1991). Experiencesofenvyand jealousy. InP. Salovey (Ed.), Thepsychology
of jealousy and envy (pp. 3-30). New York: Guilford.
Duffy, Shaw / THE SALIERI SYNDROME 21
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

Parrott, W. G., & Smith, R. H. (1993). Distinguishing the experience of envy and jealousy.
Journal of Personality and Social Psychology, 64, 906-919.
Pedhazur, E. J. (1982). Multiple regression in behavioral research:Explanation and predic
-
tion. Fort Worth, TX: Harcourt Brace.
Pushkin,A.S., &Whittaker,C.H.(1984).AlexanderPushkin:Epigramsandsatiricalverse.
Ann Arbor, MI: Ardis.
Reizenstein, R. M., & Burke, M. J. (1996,April). Another look atthe relationships between
group cohesion and groupperformance. Paperpresentedattheannualconferenceofthe
Society for Industrial and Organizational Psychology, San Diego, CA.
Robinson,S. L.,&Bennett,R.J.(1995).A typologyof deviantworkplacebehaviors:Amul
-
tidimensional scaling study. Academy of Management Journal, 38, 555-572.
Saavedra, R., &Kwun, S. K.(1993).Peer evaluation in self-managingworkgroups. Journal
of Applied Psychology, 78, 450-462.
Salovey, P. & Rodin, J. (1986). Coping with envy and jealousy. Journal of Social and Clini
-
cal Psychology, 50, 1100-1112.
Salovey, P., & Rothman, J. (1991). Envy and jealousy: Self and society. In P. Salovey (Ed.),
The psychology of jealousy and envy (pp. 280-290). New York: Guilford Press.
Seers, A. (1989). Team-member exchange quality: A new construct for role-making
research. Organizational Behavior and Human Decision Processes, 43, 118-135.
Shaw, J. D., Duffy, M. K., Jenkins, G. D., Jr., & Gupta, N. (1999). Positive and negative
affect, signal sensitivity, and pay satisfaction. Journal of Management, 25, 189-205.
Shea, G. P., & Guzzo, R. A. (1987). Group effectiveness: What really matters? Sloan Man-
agement Review, 3, 25-31.
Silver, M., & Sabini, M. (1982). The perception of envy. Social Psychology, 41, 105-117.
Smith, R. H., & Turner, T. J. (1996). Envy and Schadenfreude. Personality and Social Psy-
chology Bulletin, 22, 158-169.
Tesser, A., & Campbell, J. (1990). Self-definition: The impact of the relative performance
and similarity of others. Social Psychology Quarterly, 43, 341-347.
Tesser,A., &Smith, J.(1980). Someeffects oftask relevance andfriendship onhelping: You
don’t always help the one you like. Journal of Experimental Social Psychology, 16,
582-590.
Ulanov, A., & Ulanov, B. (1983). Cinderella and her sisters: The envied and the envying.
Philadelphia: Westminster.
Vecchio, R. P. (1995). It’s not easy being green: Jealousy and envy in the workplace.
Research in Personnel and Human Resources Management, 13, 201-244.
Wagner, J. A. (1995). Studies of individualism-collectivism: Effects cooperation in groups.
Academy of Management Journal, 38, 152-172.
White, G. L., & Mullen, P. E. (1989). Jealousy: Theory, research, and clinical strategies.
New York: Guilford.
Wolfe, J., &Box,T. M.(1988).Team cohesion effectsof businessgame performance.Simu
-
lation and Games, 19, 82-98.
Michelle K. Duffy is an assistant professor in the Gatton College of Business and
Economicsatthe University ofKentucky. She received her Ph.D.from theUniversity
of Arkansas. Her research interests include employee stress and well-being, social
undermining, and workplace deviance.
22 SMALL GROUP RESEARCH / February 2000
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from

Jason D. Shaw is an assistant professor in the Gatton College of Business and Eco
-
nomics at the University of Kentucky. He received his Ph.D. from the University of
Arkansas. His research interests include behavioral and organizational conse
-
quences of reward systems and person-environment congruence issues.
Duffy, Shaw / THE SALIERI SYNDROME 23
at The Hong Kong Polytechnic University on August 6, 2014sgr.sagepub.comDownloaded from