RESEARCH REPORT When Do Bad Apples Not Spoil the Barrel? Negative Relationships in Teams, Team Performance, and Buffering Mechanisms Jeroen P. de Jong Open University of the Netherlands Petru L. Curşeu and Roger Th. A. J. Leenders Tilburg University The study of negative relationships in teams has primarily focused on the impact of negative relationships on individual team member attitudes and performance in teams. The mechanisms and contingencies that can buffer against the damaging effects of negative relationships on team performance have received limited attention. Building on social interdependence theory and the multilevel model of team motivation, we examine in a sample of 73 work teams the team-level attributes that foster the promotive social interaction that can neutralize the adverse effect of negative relationships on team cohesion and, consequently, on team performance. The results indicate that high levels of team–member exchange as well as high task-interdependence attenuate how team cohesion and team performance suffer from negative relationships. Implications for research and practice are discussed. Keywords: negative relationships, team cohesion, team performance, social interdependence theory, motivation, multilevel Negative relationships and their associated behaviors have been suggested to harm a team from the inside (e.g., Duffy & Lee, 2012; Lam, Van der Vegt, Walter, & Huang, 2011), as they reflect negative attitudes, judgments, and behavioral intentions held by team members toward their fellow teammates (Labianca & Brass, 2006), and they are often associated with emotional distress, withdrawal, and anger (Xia, Yuan, & Gay, 2009). To date, re- search on negative relations in teams has largely built on negativity bias theory (Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001) and affective events theory (Weiss & Cropanzano, 1996)—in particular, negative affective tone (Cole, Walter, & Bruch, 2008)—focusing on the potentially damaging effects of negative relationships on work-related outcomes (e.g., Brass, Galaskiewics, Greve, & Tsai, 2004; Felps, Mitchell, & Byington, 2006; Labianca & Brass, 2006). In this article, we draw on insights from social interdepen- dence theory (SIT; Deutsch, 1949; D. W. Johnson, 2003; D. W. Johnson & Johnson, 2005a) to argue that there are structural and social team-level attributes that buffer against the detri- ments of interpersonal negative relationships in teams. Central to SIT is the idea that collaborative (generally termed promo- tive) interaction among team members is likely to enhance joint goal achievement and that obstructive (generally termed con- trient) interaction will allow some members to reach individual goals at the expense of others, while simultaneously damaging overall team-level performance. An attractive feature of SIT is that it explicitly addresses positive and negative interaction in groups (which few theories explicitly do), is very well-tested, and provid ...

1. RESEARCH REPORT
When Do Bad Apples Not Spoil the Barrel? Negative
Relationships in
Teams, Team Performance, and Buffering Mechanisms
Jeroen P. de Jong
Open University of the Netherlands
Petru L. Curşeu and Roger Th. A. J. Leenders
Tilburg University
The study of negative relationships in teams has primarily
focused on the impact of negative
relationships on individual team member attitudes and
performance in teams. The mechanisms and
contingencies that can buffer against the damaging effects of
negative relationships on team
performance have received limited attention. Building on social
interdependence theory and the
multilevel model of team motivation, we examine in a sample of
73 work teams the team-level
attributes that foster the promotive social interaction that can
neutralize the adverse effect of
negative relationships on team cohesion and, consequently, on
team performance. The results
indicate that high levels of team–member exchange as well as
high task-interdependence attenuate
how team cohesion and team performance suffer from negative
relationships. Implications for
research and practice are discussed.

2. Keywords: negative relationships, team cohesion, team
performance, social interdependence theory,
motivation, multilevel
Negative relationships and their associated behaviors have been
suggested to harm a team from the inside (e.g., Duffy & Lee,
2012;
Lam, Van der Vegt, Walter, & Huang, 2011), as they reflect
negative attitudes, judgments, and behavioral intentions held by
team members toward their fellow teammates (Labianca &
Brass,
2006), and they are often associated with emotional distress,
withdrawal, and anger (Xia, Yuan, & Gay, 2009). To date, re-
search on negative relations in teams has largely built on
negativity
bias theory (Baumeister, Bratslavsky, Finkenauer, & Vohs,
2001)
and affective events theory (Weiss & Cropanzano, 1996)—in
particular, negative affective tone (Cole, Walter, & Bruch,
2008)—focusing on the potentially damaging effects of negative
relationships on work-related outcomes (e.g., Brass,
Galaskiewics,
Greve, & Tsai, 2004; Felps, Mitchell, & Byington, 2006;
Labianca
& Brass, 2006).
In this article, we draw on insights from social interdepen-
dence theory (SIT; Deutsch, 1949; D. W. Johnson, 2003; D. W.
Johnson & Johnson, 2005a) to argue that there are structural
and social team-level attributes that buffer against the detri-
ments of interpersonal negative relationships in teams. Central
to SIT is the idea that collaborative (generally termed promo-
tive) interaction among team members is likely to enhance joint
goal achievement and that obstructive (generally termed con-
trient) interaction will allow some members to reach individual

3. goals at the expense of others, while simultaneously damaging
overall team-level performance. An attractive feature of SIT is
that it explicitly addresses positive and negative interaction in
groups (which few theories explicitly do), is very well-tested,
and provides a conceptual model on how group performance is
affected differentially by either type of interaction. Although
attractive for studying the effects of negative relationships in
teams, SIT lacks an unambiguous focus on the exact cross-level
mechanisms that motivate the team as a whole to respond to the
negative relationships between some of its members. We build
on the multilevel team motivation model (Chen & Kanfer,
2006; Chen, Kanfer, DeShon, Mathieu, & Kozlowski, 2009) to
argue that negative relationships reflect contrient (i.e., nega-
tive) interactions at the dyad level that are discretionary
(de)motivational inputs that can spiral up from the dyad to the
team level, reducing cohesion and, ultimately, team
performance.
In addition, we propose that teams possess top-down structural
and
social devices (communication intensity, team–member
exchange,
and task-interdependence) that motivate the promotive
interaction
of team members that can buffer against the otherwise damaging
effects of negative relationships.
In this article, we respond to a call for more research on the
simultaneous influence of negative and positive social
interactions
This article was published Online First March 24, 2014.
Jeroen P. de Jong, Department of Organisation, Faculty of
Management,
Science, and Technology, Open University of the Netherlands;
Petru L.

4. Curşeu and Roger Th. A. J. Leenders, Department of
Organization Studies,
School of Social and Behavioral Sciences, Tilburg University.
We thank Hans van Dijk, Will Felps, and Jing Han for their
valuable
comments on previous versions of this article.
Correspondence concerning this article should be addressed to
Jeroen P.
de Jong, Department of Organisation, Faculty of Management,
Science,
and Technology, Open University of the Netherlands,
Valkenburgerweg
177, P.O. Box 2960, 6401 DL Heerlen, the Netherlands. E-mail:
jeroen
[email protected]
T
hi
s
do
cu
m
en
t
is
co
py
ri
gh

5. te
d
by
th
e
A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
ti
on
or

6. on
e
of
it
s
al
li
ed
pu
bl
is
he
rs
.
T
hi
s
ar
ti
cl
e
is
in
te

7. nd
ed
so
le
ly
fo
r
th
e
pe
rs
on
al
us
e
of
th
e
in
di
vi
du
al
us

8. er
an
d
is
no
t
to
be
di
ss
em
in
at
ed
br
oa
dl
y.
Journal of Applied Psychology © 2014 American Psychological
Association
2014, Vol. 99, No. 3, 514 –522 0021-9010/14/$12.00 DOI:
10.1037/a0036284
514
mailto:[email protected]

9. mailto:[email protected]
http://dx.doi.org/10.1037/a0036284
on team outcomes (e.g., Chen & Sharma, 2012). Combining in-
sight from SIT with mechanisms from the multilevel team moti-
vation models is both theoretically appealing (as it provides the
foundation for the buffering role of promotive processes in
teams)
and practically relevant (as positive and negative relations
coexist
in many real life teams).
Hypotheses
Effect of Negative Relationships on Team Performance
The human cognitive system is more sensitive to stimuli that
carry negative rather than positive connotations (Baumeister,
Brat-
slavsky, Finkenauer, & Vohs, 2001); therefore, negative
relation-
ships can have more powerful consequences than positive ones
(Felps, Mitchell, & Byington, 2006; Labianca & Brass, 2006).
SIT
(Deutsch, 1949; D. W. Johnson & Johnson, 2005a) suggests that
the presence of negative relationships in a team gives rise to
so-called contrient social interaction, where team members ob-
struct each other’s task-related efforts or create an unpleasant
and
dysfunctional team atmosphere. In line with the multilevel team
motivation model of Chen and Kanfer (2006), we propose that
the
damaging effects of local negative relationships in a team
trickle
upward and affect team performance by reducing the social

10. cohe-
sion at the team level. This bottom-up process can occur
through
emotional and behavioral mechanisms.
Coworkers experience negative emotions when they witness
unpleasant interactions between others at work (Totterdell et
al.,
2012). These emotions are shared and transferred through emo-
tional contagion, vicarious affective transference, and
entrainment
(individual emotions co-evolve to achieve collective emotional
synergy), creating a negative emotional climate at the team
level
(Barsade, 2002), which disrupts team cohesion and,
consequently,
team performance. In addition, negative relationships can
induce
dysfunctional and antagonistic behaviors such as withholding of
critical information or sabotage (Chiaburu & Harrison, 2008;
Felps, Mitchell, & Byington, 2006; Lyons & Scott, 2012;
Venkat-
aramani & Dalal, 2007), and an interpersonal relation marked
by
behaviors of animosity can induce team-level relationship
conflict
(Chen, Sharma, Edinger, Shapiro, & Farh, 2011) and
consequently
reduce team cohesion. Overall, cohesion is a bottom-up
emergent
phenomenon that results from the interpersonal interactions
within
groups (Kozlowski & Chao, 2012).
According to the IMOI (input-mediator-output-input) frame-
work of team effectiveness (Ilgen, Hollenbeck, Johnson, &

11. Jundt,
2005), cohesion is factor for team effectiveness. In concert with
this, empirical studies of SIT generally find that the team mem-
bers’ tendency to be united in trying to reach joint goals has a
strong positive effect on team performance, and teams that lack
such a tendency tend to display low performance levels (for a
review, see D. W. Johnson & Johnson, 2005a). We therefore
propose that the dysfunctional interpersonal behaviors that are
associated with negative relations in a team compromise the
team’s social cohesion and, in turn, diminish team performance.
Hypothesis 1: The presence of negative relationships within a
team negatively affects team performance through its negative
effect on team cohesion.
Buffering Negative Relationships: Three Moderators
In this study, we propose that teams possess social and
structural
attributes that can buffer against the effects hypothesized above.
In
particular, we focus on three such attributes (communication
den-
sity, team member exchange, task-interdependence) that can po-
tentially attenuate the damaging effects of negative ties on the
team’s social cohesion. In SIT, it is argued that teams that focus
on
promotive interaction outperform teams that rely on contrient
interaction; a key example in this literature is that teams that
actively organize and evaluate feedback on the effectiveness of
their (joint) work processes and give preferentiality to those
pro-
cesses that turn out to be productive to the team tend to
outperform
teams that never take such evaluative efforts. This typical
example

12. of promotive interaction is in concert with literature on team
motivation (Chen & Kanfer, 2006) that emphasizes the
fundamen-
tal role of feedback as a motivating team-level factor. Although
the
predictions derived from SIT concerning the beneficial
outcomes
of promotive interactions are clear, the theory remains rather
ambiguous as to which factors are conducive to promotive inter-
action.
We build on the multilevel model of team motivation (Chen &
Kanfer, 2006; Chen et al., 2011) to propose three proximal
team-
level antecedents that motivate or enable the positive,
promotive
interaction that can attenuate the damaging influence of
negative
relationships. As for structural antecedents, we consider the
den-
sity of the communication network within the team (which cap-
tures the intensity and connectedness by which team members
interact with one another on work-related matters) and the task-
interdependence of the team members. The social antecedent we
consider is the quality of the team’s member exchange
processes.
As we argue, these three attributes reflect top-down mechanisms
that can buffer the team as a whole against local negative ties
and
can safeguard the team’s social cohesion. With social cohesion
intact, the team should retain its ability to perform well despite
the
presence of negative relationships (for the full model, see
Figure
1).

13. Negative Relationships, Cohesion, and
Communication Density
In line with the multilevel model of team motivation (Chen &
Kanfer, 2006), we argue that negative ties are discretionary
Nega�ve
rela�onships
Performance Cohesion
Team-member
exchange
Communica�on
density
Task-
interdependence
Figure 1. How teams buffer themselves against the effects of
negative
relationships.
T
hi
s
do
cu
m
en
t

14. is
co
py
ri
gh
te
d
by
th
e
A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss

15. oc
ia
ti
on
or
on
e
of
it
s
al
li
ed
pu
bl
is
he
rs
.
T
hi
s
ar
ti

16. cl
e
is
in
te
nd
ed
so
le
ly
fo
r
th
e
pe
rs
on
al
us
e
of
th
e
in

17. di
vi
du
al
us
er
an
d
is
no
t
to
be
di
ss
em
in
at
ed
br
oa
dl
y.
515NEGATIVE RELATIONSHIPS AND TEAM

18. PERFORMANCE
(de)motivational forces, likely to influence individual members
and the team as a whole. As reported in Chen, Sharma, Edinger,
Shapiro, and Farh (2011), a single interpersonal relation marked
by
contentious and personal disagreements and attacks generates
sub-
stantial relationship conflict at the team level and, as such, acts
as
a demotivating force. Nevertheless, communication density (the
extent to which team members are in mutual communication
with
each other) may buffer the (de)motivating effects of these
discre-
tionary inputs on the team as a whole.
When all team members are involved in frequent, mutual con-
versations regarding the team task or informal issues unrelated
to
their work, the team has access to channels of interaction that
allow it to quickly respond to unproductive events. The commu-
nication interconnectedness is thought to influence the amount
of
information and support that can move through the team
(Balkundi
& Harrison, 2006). When communication density is high, team
members regularly communicate with each other and are better
able to deal with the destructive effects of negative
interpersonal
relations. Although there may be conditions under which
intercon-
nectedness allows for negativity to become contagious, we
argue

19. that a team with well-connected members will be able to fence
the
negativity locally (i.e., block it from spiraling up to the team
level)
and exhibit enough elasticity to keep its cohesion intact. Dense
positive exchanges in teams, instrumental or informal,
positively
relate to members’ attitudes and well-being (LePine, Methot,
Crawford, & Buckman, 2012); we contend that teams with
denser
communication networks are better able to handle negative
social
exchanges in the team when such exchanges emerge. Frequent
interaction in a team provides those team members who are in-
volved in negative relationships with positive motivational
expe-
riences that soften the impact of the stressful negative
experience
(Cohen & Wills, 1985) and can help them to come to grips with
their negative experiences. Teams with high communication
den-
sity provide their members with the opportunity to directly con-
tribute to an atmosphere in which the negative relations become
much less harmful. In addition, teams with high communication
density leave little opportunity for negativity to spread through
the
team because communication density positively impacts the
team’s capacity to coordinate its actions (Reagans &
Zuckerman,
2001). Decreased team density has been suggested to make it
more
difficult for the team to maintain the closure needed for
effective
positive norms and sanctions, which is an additional way in
which
the potentially destructive behavior that can follow negative

20. rela-
tionships can be controlled (Brass & Labianca, 1999). In sum,
we
propose that communication density is an effective team-level
structural device that provides a buffer against the potentially
adverse effects of negative relationships.
Hypothesis 2: The effect of negative relationships on team
cohesion is moderated by team communication density: the
higher the communication density, the lower the damaging
effect of negative relationships on team cohesion. As a result,
high communication density neutralizes the damaging effect
of negative relationships on team performance.
Negative Relationships, Cohesion, and Task-
Interdependence
A second team-level structural attribute that can attenuate the
undesirable effects of negative relationships is the extent to
which
team members are task-interdependent. Task-interdependence
re-
flects the extent to which collective performance depends on
successful intrateam information sharing, effective
coordination,
and mutual adjustment (Saavedra, Earley, & Vandyne, 1993).
When task-interdependence is high, team members depend on
each other to complete their collective task: They sink or swim
together; that is, they have to maximize their own effectiveness
and maximize the effectiveness of all other group members
(D. W. Johnson & Johnson, 2005b). In other words, task-
interdependence facilitates top-down motivational processes as
highly task-interdependent teams rely on continuous internal
mon-
itoring to cope with stressors that interfere with task completion

21. (Kozlowski, Gully, Nason, & Smith, 1999). This monitoring
sys-
tem decreases the leeway team members have to engage in the
dysfunctional behaviors that result from negative relationships
(e.g., shirking and free riding; Jones, 1984). Conversely, teams
engaged in tasks that require a low degree of task-
interdependence
are unlikely to develop an internal monitoring system and team
members may, as a result, engage in dysfunctional behaviors in
response to negative relationships. To summarize, task-
interdepen-
dence—a team-level attribute—motivates the promotive interac-
tion (at the dyadic level) in teams that can mitigate the adverse
consequences of negative relationships.
Hypothesis 3: The effect of negative relationships on team
cohesion is moderated by task-interdependence: the higher the
task-interdependence, the lower the damaging effect of nega-
tive relationships on team cohesion. As a result, high task-
interdependence neutralizes the damaging effect of negative
relationships on team performance.
Negative Relationships, Cohesion, and the Quality of
Social Exchange
Because team members are part of a social system of interacting
roles, the quality of their social exchange can provide a team-
level
buffer that reduces the adverse consequences of negative
relation-
ships (Duffy & Lee, 2012; Seers, 1989). Team–member
exchange
(TMX) refers to “the reciprocity between a member and his or
her
team with respect to the member=s contribution of ideas,
feedback,

22. and assistance to other members and, in turn, the member=s
receipt
of information, help, and recognition from other team members”
(Seers, Petty, & Cashman, 1995, p. 21). Focusing on the
exchange
of team members with the team as a whole, TMX considers the
team (rather than specific team members) as the agent for ex-
change quality. Seeking and receiving support, assistance and
information is especially beneficial in stressful situations
(Carver,
Scheier, & Weintraub, 1989), such as those caused by negative
relationships. For example, high quality team-member exchange
was found to help newcomers to cope with the negative effects
of
unmet expectations (Major, Kozlowski, Chao, & Gardner,
1995).
Team members who are involved in negative relationships find
support by experiencing an atmosphere in which one is naturally
helped by others, feels recognized, and shares responsibility. A
positive team atmosphere is a top-down motivational force
(Chen
& Kanfer, 2006) that makes it easier for team members to deal
with negativity, because it supports one’s self-esteem and the
extent to which the team member identifies with the team and
feels
part of a cohesive whole. When members of a team develop
strong
T
hi
s
do
cu

23. m
en
t
is
co
py
ri
gh
te
d
by
th
e
A
m
er
ic
an
P
sy
ch
ol
og
ic

24. al
A
ss
oc
ia
ti
on
or
on
e
of
it
s
al
li
ed
pu
bl
is
he
rs
.
T
hi

25. s
ar
ti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r
th
e
pe
rs
on
al
us
e
of

26. th
e
in
di
vi
du
al
us
er
an
d
is
no
t
to
be
di
ss
em
in
at
ed
br
oa

27. dl
y.
516 DE JONG, CURŞEU, AND LEENDERS
norms of supporting each other, the impact of social support in
the
team may be stronger, relative to circumstances in which some
members support each other and others do not (Chen & Sharma,
2012). Such an environment is well-suited to buffering the
harmful
effects of negative relationships on cohesion and, ultimately,
per-
formance. In summary, TMX increases cross-level (top-down)
motivation in favor of promotive interaction and provides an
environment that is preeminently suited for effective evaluation
of
activities and behavior by both the individual and the group as a
whole.
Hypothesis 4: The effect of negative relationships on team
cohesion is moderated by TMX: the higher the TMX, the
lower the damaging effect of negative relationships on team
cohesion. As a result, high TMX neutralizes the damaging
effect of negative relationships on team performance.
Method
Procedure and Participants
For this study we collected data in eight organizations (two
municipalities, a bank, a web-based retail organization, a
charita-

28. ble institution, a medium-sized university, a health care
organiza-
tion, and a local police department) located in a medium-size
Western European country. Data were obtained through two
ques-
tionnaires administered to team members and the formal leader
of
the team, respectively. Each leader in our data set was
responsible
for one team only. The questionnaires were completed during
working hours, in on-site meeting rooms, with a researcher
pres-
ent. The sample consists of 334 respondents from 73 teams with
an
average within-team response rate of 84%. The average team
size,
excluding the team leader, was 4.6 (ranging from three to 13
members, SD � 2.2). The mean respondent age was 39 years
(ranging from 17 to 63 years, SD � 11.95), with 152 women
(45%) and 182 men (55%). The mean age of the leader was 42
years (SD � 9.8), with 21 women (29%) and 52 men (71%). The
majority of the leaders in our sample have a college or
university
degree (83%). Their average tenure with the organization is 98
months (SD � 122).
Measures
Presence of negative relationships. Following, among oth-
ers, Kane and Labianca (2005) and Venkataramani and Dalal
(2007), we provided team members with a list with names of
team
members (excluding the team leader) and asked them to report
their relationship with all other members of the team on a 4-
point
scale (1 � dislike a lot, 2 � dislike, 3 � like, 4 � like a lot). A

29. score of 1 or 2 indicated a negative relationship. We followed
Labianca and Brass’s (2006) suggestion that a relevant negative
relationship exists when at least one person in the relationship
has
a level of dislike for another person. About 7% of the
relationships
were negative, a percentage that conforms to previous empirical
research on negative relationships (e.g., Baldwin, Bedell, &
John-
son, 1997; Gersick, Bartunek, & Dutton, 2000; Labianca, Brass,
&
Gray, 1998). In 32 teams (44%) we found at least one negative
relationship. The number of negative relationships in a team
ranged from one to 11 (in a team of 11 members); the highest
percentage within a team was 33% (in a team of six members).
Because these high percentages were exceptional and most of
the
teams (n � 20) had only one or two negative relationship(s), we
dichotomized the variable so it reflects the presence or absence
of
negative relationships. We did, however, repeat our analyses
using
the proportion of negative relationships within the team as a
dependent variable; this yielded the same results as when using
the
dichotomous variable. We only show and discuss these latter
results.
Team cohesion. We measured team cohesion with Chang and
Bordia’s (2001) eight-item scale comprising items like “Our
team
is united in trying to reach its goal for performance.” The items
are
rated on a 5-point Likert-type scale ranging from 1 � strongly
disagree to 5 � strongly agree and cover both task and social

30. cohesion. Given the good internal consistency of the scale and
the
meta-analytical evidence (Beal, Cohen, Burke, & McLendon,
2003) showing that task and social cohesion are both positively
associated to group performance, we used the scale’s overall
score
for group cohesion in the analyses. Because our hypotheses are
formulated at the team level, individual answers were
aggregated
to the team level. To test whether such aggregation was
justified,
we calculated a within-group interrater reliability statistic,
rwg(j)
(James, Demaree, & Wolf, 1993), and intraclass correlation
indi-
ces ICC(1) and ICC(2) (Bliese & Halverson, 1998). Because the
distribution of the cohesion scale is slightly skewed, we used an
expected error variance of 1.34 (Lebreton & Senter, 2008).1 The
mean rwg(j) for cohesion was .92 (median � .93, range � .78 to
1.00). Furthermore, ICC(1) was .22 and ICC(2) was .56, F �
2.372, p � .05.
Communication density. Communication density refers to
the extent to which team members are in communication with
each
other regarding the team task and/or informal topics. We asked
respondents to indicate how frequently they communicate with
the
other members of the team “about their work for the team
and/or
about informal topics not related to the task” on the following
scale: 0 � never, 1 � less than once a month, 2 � 1 to 3 times a
month, 3 � 1 to 3 times a week, and 4 � daily. We computed
density by dividing the reported frequency of interactions of all
team members by the possible total frequency of interactions in
the

31. team (Sparrowe, Liden, Wayne, & Kraimer, 2001).
TMX. We measured TMX with the Seers, Petty, and Cashman
(1995) 10-item scale, consisting of statements such as “My team
members understand my problems and needs.” The items are
rated
on a 5-point Likert-type scale ranging from 1 � strongly
disagree
to 5 � strongly agree. The mean rwg(j) is .94 (Mdn � .95, range
�
.83 to 1.00), ICC(1) � .28 and ICC(2) � .68, F � 2.725, p �
.05,
justifying aggregation of the individual-level data to the team
level.
1 Recently, concerns have been voiced regarding the habitual
reliance on
uniform null distributions for rwg(j) (e.g., Lebreton & Senter,
2008) in most
articles. Because our data are skewed slightly, we used the
expected error
variance of 1.34 Lebreton and Senter (2008, Table 2) suggested
for this
level of skew. If we had used the uniform distribution as our
null distri-
bution, the rwg(j)’s would have become inflated. To see this,
the rwg(j)’s
under a uniform null distribution (with variance 2) are indeed
higher: .96
(Mdn � .96, range � .90 to 1.00) for cohesion, .96 (Mdn � .97,
range �
.90 to 1.00) for TMX and .78 (Mdn � .79, range � .40 to 1.00)
for
task-interdependence.
T

32. hi
s
do
cu
m
en
t
is
co
py
ri
gh
te
d
by
th
e
A
m
er
ic
an
P
sy

33. ch
ol
og
ic
al
A
ss
oc
ia
ti
on
or
on
e
of
it
s
al
li
ed
pu
bl
is
he

34. rs
.
T
hi
s
ar
ti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r
th
e
pe
rs
on

35. al
us
e
of
th
e
in
di
vi
du
al
us
er
an
d
is
no
t
to
be
di
ss
em
in

36. at
ed
br
oa
dl
y.
517NEGATIVE RELATIONSHIPS AND TEAM
PERFORMANCE
Task-interdependence. We measure task-interdependence
using the six-item scale by Van der Vegt, Emans, and Van de
Vliert (2001) containing items such as “I depend on my
colleagues
for the completion of my work” rated on a scale of 1 � disagree
strongly to 5 � agree strongly. The mean rwg(j) is .67 (Mdn �
.68,
range � .23 to 1.00), ICC(1) � .16, and ICC(2) � .46, F �
1.864,
p � .05, justifying aggregation of the data to the team level.
Team performance. The team leader evaluated team perfor-
mance using the scale developed by Rousseau and Aubé (2010):
the team leader is asked to evaluate the team’s outcomes over
the
recent 6 months on five different performance criteria: (a)
achieve-
ment of performance goals, (b) productivity (quantity of work),
(c)
quality of work accomplished, (d) respect for deadlines, and (e)
respect for costs. Team performance was rated on a 5-point

37. scale
ranging from 1 � very low to 5 � very high.
Control variables. Team size is used as a control variable
because team size has been found to relate negatively to
cohesion
(Carron & Spink, 1995). We also included average team tenure
(in
months) as a control variable. High team tenure can give rise to
teams relying on routine and using cognitive short-cuts to avoid
conflict (Katz, 1982). This may increase team cohesion. In a
separate analysis, we accounted for possible organization-level
differences by adding organization dummies to the regressions.
Because the addition of these dummies did not change the
results,
they are not reported below.
To test for discriminant validity of TMX, task-interdependence,
and cohesion, we conducted a confirmatory factor analysis
using
three different models (communication density was operational-
ized as network density, making confirmatory factor analysis ir-
relevant for it). We tested a one-factor model including all three
variables simultaneously, �2(209) � 1216.72, nonnormed fit
index
[NNFI] � .83, comparative fit index [CFI] � .86, root-mean-
square error of approximation [RMSEA] � .11; a two-factor
model in which TMX and cohesion were captured in one latent
factor, �2(208) � 928.82, NNFI � .87, CFI � .90, RMSEA �
.09;
and a three-factor model in which each factor represents the
latent
construct of each variable, �2(206) � 661.42, NNFI � .91, CFI
�
.94, RMSEA � .07. The confirmatory factor analyses show that
the three-factor model fits the data best, supporting the

38. inclusion of
the three constructs as separate variables in the analysis.
Analyses
To test Hypothesis 1, we used the procedure recommended by
Preacher and Hayes (2004) to analyze if cohesion mediates the
relationship between the presence of negative relationships and
team performance. Hypotheses 2, 3, and 4 propose a mediated
moderation model. To test for this mediated moderation, we ap-
plied the procedure proposed by Preacher, Rucker, and Hayes
(2007), testing this indirect association at three levels of
commu-
nication density, TMX, and task-interdependence (�1 standard
deviation, mean, 1 standard deviation). All variables were
entered
as continuous variables, which we mean-centered before
proceed-
ing with the analyses. Bias-corrected bootstrapping with 5,000
resamples was applied to test whether the indirect associations
differ significantly from zero.
Results
Table 1 shows the descriptive statistics, correlations, and Cron-
bach’s alphas of the main variables used in this study. In testing
Hypothesis 1, we regressed the cohesion on the presence of neg-
ative relationships (together with the control variables) first and
then regressed team performance on the presence of negative
relationships together with cohesion in a second regression (see
Table 2). The results show that the presence of negative
relation-
ships is negatively related to cohesion (B � �.39, p � .05),
which,
in turn, is positively related to team performance (B � .53, p �
.05), while the presence of negative relationships appears to be

39. unrelated to team performance (B � �.22, ns). The
bootstrapping
results reveal an indirect association between the presence of
negative relationships and team performance mediated by cohe-
sion (effect size � �.18, 95% confidence interval [95% CI]
[�.37, �.06]), which supports Hypothesis 1.2
In Table 2, we also present the results of the regressions used to
test Hypotheses 2, 3, and 4. We include communication density,
task-interdependence, TMX, and their interaction in the model
with negative relationships to test their association with team
cohesion.3 For communication density, neither the main effect
(B � �.12, ns) nor its interaction (B � �.16, ns) with negative
relationships is statistically significant; we therefore reject Hy-
pothesis 2. For both task-interdependence (B � .35, p � .05)
and
TMX (B � .72, p � .05), the interaction with negative relation-
ships is statistically significant. Figure 2A shows that at high
levels
of task-interdependence (�1 standard deviation), the association
between presence of negative relationships and cohesion no
longer
appears. Figure 2B paints a similar picture for TMX: At low
levels
of TMX (�1 standard deviation), negative relationships are neg-
atively associated with cohesion, but at high levels of TMX (�1
standard deviation), the association disappears. In line with Hy-
potheses 3 and 4, this shows that high task-interdependence and
TMX are effective in neutralizing the adverse effects of
negative
relationships on team cohesion.
We used bootstrapping (Preacher et al., 2007) to further show
how the indirect association between negative relationships and
team performance is mediated by cohesion at different levels of
task-interdependence and TMX. The negative indirect

40. association
appears to be statistically significant at medium (effect
size � �.08, 95% CI [�.23, �.01]) and especially at low levels
of
2 To assess whether cohesion fully mediates between the
presence of
negative relationships and team performance, we applied the
criteria for
full mediation proposed by Mathieu and Taylor (2006).
Following their
decision tree (Mathieu & Taylor, 2006, p. 1040), we established
that,
controlling for team size and average team tenure, the presence
of negative
relationships is associated with team performance, Byx � �.42
(SE � .16),
p � .05; that the presence of negative relationships is associated
with
cohesion, Bmx � �.39 (SE � .09), p � .05; and that cohesion is
related to
team performance, Bym � .53 (SE � .20), p � .05. Finally, we
find a
nonsignificant association of the relationship between the
presence of
negative relationships and team performance while controlling
for cohe-
sion, Byx, m � �.22 (SE � .17), ns, which meets the criteria to
establish full
mediation as proposed by Mathieu and Taylor (2006).
3 To test if the moderators also moderate elsewhere in the
model, we
performed additional regressions to test if the moderation is
significant in
the relationship between (a) negative relationships and team

41. performance
and (b) cohesion and team performance. The results of the
regressions
show that this is not the case; moderation is only present in the
relationship
between negative relationships and cohesion.
T
hi
s
do
cu
m
en
t
is
co
py
ri
gh
te
d
by
th
e
A
m

42. er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
ti
on
or
on
e
of
it
s
al
li

43. ed
pu
bl
is
he
rs
.
T
hi
s
ar
ti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r

44. th
e
pe
rs
on
al
us
e
of
th
e
in
di
vi
du
al
us
er
an
d
is
no
t
to

45. be
di
ss
em
in
at
ed
br
oa
dl
y.
518 DE JONG, CURŞEU, AND LEENDERS
task-interdependence (�1 standard deviation, effect size �
�.17,
95% CI [�.39, �.02]), although it fades away at a high level of
task-interdependence (�1 standard deviation, effect size � .01,
95% CI [�.12, .14]). We find similar results for TMX as a
moderator. The negative indirect association is statistically
signif-
icant at medium (effect size � �.09, 95% CI [�.25, �.01]) and
low levels of TMX (�1 standard deviation, effect size � �.20,
95% CI [�.44, �.06]) but not at a high level of TMX (�1
standard
deviation, effect size � .01, 95% CI [�.15, .20]). These results
indicate that negative relationships damage team performance
through their adverse effect on team cohesion in teams with low

46. or
intermediate levels of task-interdependence and TMX. However,
in teams that are high on task-interdependence or TMX—teams
that have strong mechanisms that foster promotive interaction—
negative relationships no longer have an adverse effect on team
cohesion and, consequently, team performance remains intact.
This supports Hypotheses 3 and 4.
Discussion
Our study provides initial insight regarding the conditions under
which teams do or do not suffer from negative interpersonal
relationships. Consistent with our first hypothesis, we found
that
negative relationships reduce team cohesion, which, in turn, de-
creases team performance. Given that team cohesion is one of
the
main antecedents of team performance (Chang & Bordia, 2001),
Table 1
Means, Standard Deviations, and Correlations of the Main
Variables
Variable M SD 1 2 3 4 5 6 7 8
1. Negative relationshipsa .44 .50 —
2. Team cohesion 3.64 0.38 �.50� (.84)
3. Team performance 3.45 0.61 �.27� .36� (.82)
4. Communication density .73 .13 �.32� .25� .18 —
5. Task-interdependence 3.29 0.44 �.14 .23� �.01 .18 (.75)
6. TMX 3.84 0.23 �.43� .64� .19 .25� .27� (.84)
7. Team size 4.55 2.20 .45� �.26� �.02 �.17 .01 �.25� —
8. Average team tenureb 49.59 36.10 .29� �.22 �.08 �.11
�.26� �.19 .58� —

47. Note. N � 73. Values in parentheses represent Cronbach’s
alphas, which reflect the internal consistency reliability of the
(aggregated) team-level
measures. TMX � team–member exchange.
a This is a binary value for which 0 � no negative relationships
present and 1 � negative relationship present. b Tenure is given
in months.
� p � .05.
Table 2
Results of the Ordinary Least Squares Regression Analyses (N
� 73) Used to Test the Association Between Presence of
Negative
Relationships and Team Performance Through Team Cohesion
(Hypothesis 1) at Different Levels of Communication Density
(Hypothesis 2), Task-Interdependence (Hypothesis 3), and TMX
(Hypothesis 4)
Variable
Team cohesion Team performance
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 1
Model 2
B SE B SE B SE B SE B SE B SE B SE B SE
Control variables
Team size �.01 .02 .01 .02 �.01 .02 �.01 .02 .00 .02 �.02 .02
.08 .04 .08 .04
Average team tenure .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00
.00 .00 .00 .00 .00
Main effects
Negative relationshipsa �.39� .09 �.25� .09 �.24� .09 �.20�
.08 �.18� .08 �.16 .08 �.42� .16 �.22 .17

48. Team cohesion .53� .20
Moderators
Communication density �.26 .27 �.30 .29 �.08 .27 �.26 .26
�.12 .28
Task-interdependence .06 .08 .05 .09 .07 .08 .08 .08 .08 .08
TMX .79� .17 .79� .17 .85� .16 .89� .17 .93� .16
Interactions
Negative Relationships �
Communication Density �.27 .61 �.16 .58
Negative Relationships �
Task-Interdependence .40� .16 .35� .16
Negative Relationships � TMX .83� .35 .72� .34
R2 .28 .49 .49 .54 .54 .57 .11 .19
R2-changeb .21� .00 .05� .05� .08� .09� .08�
Note. TMX � team–member exchange.
a This is a binary value for which 0 � no negative relationships
present and 1 � negative relationship present. b The values of
R2-change for Models 3,
4, 5 and 6 reflect the comparison with Model 2.
� p � .05.
T
hi
s
do
cu
m

49. en
t
is
co
py
ri
gh
te
d
by
th
e
A
m
er
ic
an
P
sy
ch
ol
og
ic
al

50. A
ss
oc
ia
ti
on
or
on
e
of
it
s
al
li
ed
pu
bl
is
he
rs
.
T
hi
s

51. ar
ti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r
th
e
pe
rs
on
al
us
e
of
th

52. e
in
di
vi
du
al
us
er
an
d
is
no
t
to
be
di
ss
em
in
at
ed
br
oa
dl

53. y.
519NEGATIVE RELATIONSHIPS AND TEAM
PERFORMANCE
we focused the main part of this study on team-level factors that
can motivate promotive, positive interaction in teams and, as
proximal antecedents of such positive processes, buffer the team
against the detrimental influence of negative relations.
We found support for our hypotheses that task-interdependence
and TMX act as such buffers against the adverse effect of
negative
relationships. The analysis shows that negative relationships
clearly harm team cohesion in those teams that are low on task-
interdependence or TMX—these are teams in which members do
not feel helped and respected and in which members barely
depend
on others to perform their jobs. In contrast, teams that perform
well
despite having negative relations are characterized by high-
quality
interpersonal interaction and above-average levels of task-
interdependence that retain a harmonious, goal-directed social
climate within the team. In other words, in line with the
multilevel
team motivation models (Chen & Kanfer, 2006), we show that
task-interdependence and TMX act as beneficial top-down
devices
that motivate team members to engage in the type of promotive
interaction that can effectively buffer against the detrimental ef-
fects of negative relationships. In teams where these devices are
lacking, such promotive interaction is not stimulated, and the
negative relationships retain their full damaging effect on team

54. cohesion.
We did not find support for communication density to create a
buffer against negative relationships. It may be, however, that
communication density plays a somewhat different role with re-
spect to negative relationships: The social skills that underlie
communication density could directly decrease the probability
of
the occurrence of negative relationships in the first place
(LePine
et al., 2012). Task-interdependence and TMX are top-down
moti-
vational forces that explicitly encourage (group) processing
activ-
ity and ultimately neutralize the perils of contrient interactions
in
teams. The positive and critical activity of group processing
helps
team members solve the cognitive dissonance inherent to situa-
tions when negative relationships are present and task-
interdependence is high (group members having to work
together
while experiencing negative interpersonal relations at work).
Em-
pirical evidence stemming from conflict theory shows that high
levels of interdependence fosters efforts to work hard and cope
with the presence of negative relationships (Sherif, Harvey,
White,
Hood, & Sherif, 1961), a finding echoed in structural adaptation
theory (M. D. Johnson et al., 2006), a dynamic extension of
SIT.
This study makes several contributions to research on intrateam
relations and team performance. Adding to literature that has
argued for the impact of negative relationships on individual

55. team
members (Totterdell et al., 2012; Xia et al., 2009), our research
addresses the effect of negative relationships on the team as a
whole. Moreover, previous research on negative relationships in
teams has focused on direct (negative) effects of negative
relation-
ships on team performance (e.g., Sparrowe et al., 2001) or on
structural consequences of negative relationships (e.g., Huitsing
et
al., 2012; Hummon & Doreian, 2003). In the current study, we
combine insights from SIT and the multilevel motivation models
to specify top-down team-level attributes (i.e., two structural
and
one social attribute) that can motivate and enable promotive
inter-
action in teams. We believe this approach enriches SIT
consider-
ably. To date, empirical SIT-based research has not addressed
the
situation in which promotive and contrient interactions occur si-
multaneously within the same team; this study adds to SIT by
showing how mechanisms that foster promotive interaction in a
team (top-down motivational forces) can counteract the
damaging
effects of simultaneously existing contrient interaction (bottom-
up
discretionary demotivating factors), such that the team can still
perform well. Most people, in their lives, will be a member of at
least one team with negative relationships; combining the
insights
of multilevel team motivation models with SIT to address situa-
tions like this, therefore, is practically relevant and allows SIT
researchers to expand their theorizing to this commonly
occurring
setting.

56. Limitations and Avenues for Future Research
Our study has some limitations. First, Labianca and Brass
(2006) argued that the social liability of negative relationships
may
be determined by their strength, reciprocity, the degree of
aware-
ness, and the social distance between the actors involved. As we
investigated work teams, we expected that social distance was
low
and the degree of awareness was likely to be high. In future
studies, researchers could explore the impact of these
characteris-
tics of negative relationships. Second, as our study relies on the
presence or absence of negative relationships, it does not
address
the way in which the within-team structure of negative relation-
ships affects team cohesion. We propose that research could
move
beyond exploring structures of positive relationships (which is
the
common approach in most of social network analysis) to
exploring
Figure 2. Association between presence of negative
relationships and
team cohesion moderated by (A) task-interdependence (TI) and
(B) team–
member exchange (TMX).
T
hi
s
do

57. cu
m
en
t
is
co
py
ri
gh
te
d
by
th
e
A
m
er
ic
an
P
sy
ch
ol
og

58. ic
al
A
ss
oc
ia
ti
on
or
on
e
of
it
s
al
li
ed
pu
bl
is
he
rs
.
T

59. hi
s
ar
ti
cl
e
is
in
te
nd
ed
so
le
ly
fo
r
th
e
pe
rs
on
al
us
e

60. of
th
e
in
di
vi
du
al
us
er
an
d
is
no
t
to
be
di
ss
em
in
at
ed
br

61. oa
dl
y.
520 DE JONG, CURŞEU, AND LEENDERS
the way in which negative relationship structures impact team
dynamics and performance.4 Third, we used a cross-sectional
design and therefore we can make no causal claims. We suggest
future research to explore the impact of negative relationships
on
team dynamics and performance through a longitudinal perspec-
tive.
Conclusions
The presence of negative relationships in teams has a negative
association with team cohesion and performance. However, sev-
eral mechanisms that motivate promotive interaction appear to
be
able to attenuate this negative impact. A practically useful
obser-
vation is that several of these contingencies can be managed by
the
team or by team leaders. For example, TMX can potentially be
enhanced by stimulating helping norms in a team, and task-
interdependence is a variable open to manipulation through
active
managerial decision making. Given that only very little is
known
about the factors that fuel the commencement of negative
relation-
ships, it may be worthwhile for managers to have some handles

62. that can proactively make their teams more resilient to the
occur-
rence of negative relationships.
4 A first exploratory analysis on our data revealed that the in-
degree and
out-degree of disliked team members (within the liking or
disliking net-
work) and the in-degree and out-degree of the disliked team
members
(within the communication network) was not related to team
cohesion
(N � 32).
References
Baldwin, T. T., Bedell, M. D., & Johnson, J. L. (1997). The
social fabric
of a team-based MBA program: Network effects on student
satisfaction
and performance. Academy of Management Journal, 40, 1369 –
1397.
doi:10.2307/257037
Balkundi, P., & Harrison, D. A. (2006). Ties, leaders, and time
in teams:
Strong inference about network structure’s effects on team
viability and
performance. Academy of Management Journal, 49, 49 – 68.
doi:
10.5465/AMJ.2006.20785500
Barsade, S. G. (2002). The ripple effect: Emotional contagion
and its
influence on group behavior. Administrative Science Quarterly,
47,

63. 644 – 675. doi:10.2307/3094912
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K.
D. (2001).
Bad is stronger than good. Review of General Psychology, 5,
323–370.
doi:10.1037/1089-2680.5.4.323
Beal, D. J., Cohen, R. R., Burke, M. J., & McLendon, C. L.
(2003).
Cohesion and performance in groups: A meta-analytic
clarification of
construct relations. Journal of Applied Psychology, 88, 989 –
1004. doi:
10.1037/0021-9010.88.6.989
Bliese, P. D., & Halverson, R. R. (1998). Group size and
measures of
group-level properties: An examination of eta-squared and ICC
values.
Journal of Management, 24, 157–172. doi:10.1016/S0149-
2063(99)80058-0
Brass, D. J., Galaskiewics, J., Greve, H. R., & Tsai, W. (2004).
Taking
stock of networks and organizations: A multilevel perspective.
Academy
of Management Journal, 47, 795– 817. doi:10.2307/20159624
Brass, D. J., & Labianca, G. (1999). Social capital, social
liabilities, and
social resources management. In R. T. A. J. Leenders & S. M.
Gabbay
(Eds.), Corporate social capital and liability (pp. 323–338).
Norwell,
MA: Kluwer Academic. doi:10.1007/978-1-4615-5027-3_18

64. Carron, A. V., & Spink, K. S. (1995). The group-size cohesion
relationship
in minimal groups. Small Group Research, 26, 86 –105.
doi:10.1177/
1046496495261005
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989).
Assessing coping
strategies: A theoretically based approach. Journal of
Personality and
Social Psychology, 56, 267–283. doi:10.1037/0022-
3514.56.2.267
Chang, A., & Bordia, P. (2001). A multidimensional approach
to the group
cohesion-group performance relationship. Small Group
Research, 32,
379 – 405. doi:10.1177/104649640103200401
Chen, G., & Kanfer, R. (2006). Toward a systems theory of
motivated
behavior in work teams. Research in Organizational Behavior,
27,
223–267. doi:10.1016/S0191-3085(06)27006-0
Chen, G., Kanfer, R., DeShon, R. P., Mathieu, J. E., &
Kozlowski, S. W. J.
(2009). The motivating potential of teams: A test and extension
of Chen
and Kanfer’s (2006) model. Organizational Behavior and
Human De-
cision Processes, 110, 45–55. doi:10.1016/j.obhdp.2009.06.006
Chen, G., & Sharma, P. N. (2012). Bringing together the yin and
yang of

65. social exchanges in teams. In L. T. Eby & T. D. Allen (Eds.),
Personal
relationships: The effect on employee attitudes, behavior, and
well-being
(pp. 221–234). New York, NY: Routledge.
Chen, G., Sharma, P. N., Edinger, S., Shapiro, D. L., & Farh, J.
L. (2011).
Motivating and de-motivating forces in teams: Cross-level
influences of
empowering leadership and relationship conflict. Journal of
Applied
Psychology, 96, 541–557. doi:10.1037/a0021886
Chiaburu, D. S., & Harrison, D. A. (2008). Do peers make the
place?
Conceptual synthesis and meta-analysis of coworker effects on
percep-
tions, attitudes, OCBs, and performance. Journal of Applied
Psychology,
93, 1082–1103. doi:10.1037/0021-9010.93.5.1082
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the
buffering
hypothesis. Psychological Bulletin, 98, 310 –357.
doi:10.1037/0033-
2909.98.2.310
Cole, M. S., Walter, F., & Bruch, H. (2008). Affective
mechanisms linking
dysfunctional behavior to performance in work teams: A
moderated
mediation study. Journal of Applied Psychology, 93, 945–958.
doi:
10.1037/0021-9010.93.5.945

66. Deutsch, M. (1949). A theory of co-operation and competition.
Human
Relations, 2, 129 –152. doi:10.1177/001872674900200204
Duffy, M. K., & Lee, K. Y. (2012). Negative interpersonal
exchanges in
teams. In L. T. Eby & T. D. Allen (Eds.), Personal
relationships: The
effect on employee attitudes, behavior, and well-being (pp.
195–220).
New York, NY: Routledge.
Felps, W., Mitchell, T. R., & Byington, E. (2006). How, when,
and why
bad apples spoil the barrel: Negative group members and
dysfunctional
groups. Research in Organizational Behavior, 27, 175–222. doi:
10.1016/S0191-3085(06)27005-9
Gersick, C. J. G., Bartunek, J. M., & Dutton, J. E. (2000).
Learning from
academia: The importance of relationships in professional life.
Academy
of Management Journal, 43, 1026 –1044. doi:10.2307/1556333
Huitsing, G., van Duijn, M. A. J., Snijders, T. A. B., Wang, P.,
Sainio, M.,
Salmivalli, C., & Veenstra, R. (2012). Univariate and
multivariate mod-
els of positive and negative networks: Liking, disliking, and
bully–
victim relationships. Social Networks, 34, 645– 657.
doi:10.1016/j
.socnet.2012.08.001
Hummon, N. P., & Doreian, P. (2003). Some dynamics of social

67. balance
processes: Bringing Heider back into balance theory. Social
Networks,
25, 17– 49. doi:10.1016/S0378-8733(02)00019-9
Ilgen, D. R., Hollenbeck, J. R., Johnson, M., & Jundt, D.
(2005). Teams in
organizations: From input-process-output models to IMOI
models. An-
nual Review of Psychology, 56, 517–543.
doi:10.1146/annurev.psych.56
.091103.070250
James, L. R., Demaree, R. G., & Wolf, G. (1993). rwg: An
assessment of
within-group interrater agreement. Journal of Applied
Psychology, 78,
306 –309. doi:10.1037/0021-9010.78.2.306
T
hi
s
do
cu
m
en
t
is
co
py
ri

68. gh
te
d
by
th
e
A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
ti
on

69. or
on
e
of
it
s
al
li
ed
pu
bl
is
he
rs
.
T
hi
s
ar
ti
cl
e
is
in

70. te
nd
ed
so
le
ly
fo
r
th
e
pe
rs
on
al
us
e
of
th
e
in
di
vi
du
al

71. us
er
an
d
is
no
t
to
be
di
ss
em
in
at
ed
br
oa
dl
y.
521NEGATIVE RELATIONSHIPS AND TEAM
PERFORMANCE
http://dx.doi.org/10.2307/257037
http://dx.doi.org/10.5465/AMJ.2006.20785500
http://dx.doi.org/10.5465/AMJ.2006.20785500

72. http://dx.doi.org/10.2307/3094912
http://dx.doi.org/10.1037/1089-2680.5.4.323
http://dx.doi.org/10.1037/0021-9010.88.6.989
http://dx.doi.org/10.1037/0021-9010.88.6.989
http://dx.doi.org/10.1016/S0149-2063%2899%2980058-0
http://dx.doi.org/10.1016/S0149-2063%2899%2980058-0
http://dx.doi.org/10.2307/20159624
http://dx.doi.org/10.1007/978-1-4615-5027-3_18
http://dx.doi.org/10.1177/1046496495261005
http://dx.doi.org/10.1177/1046496495261005
http://dx.doi.org/10.1037/0022-3514.56.2.267
http://dx.doi.org/10.1177/104649640103200401
http://dx.doi.org/10.1016/S0191-3085%2806%2927006-0
http://dx.doi.org/10.1016/j.obhdp.2009.06.006
http://dx.doi.org/10.1037/a0021886
http://dx.doi.org/10.1037/0021-9010.93.5.1082
http://dx.doi.org/10.1037/0033-2909.98.2.310
http://dx.doi.org/10.1037/0033-2909.98.2.310
http://dx.doi.org/10.1037/0021-9010.93.5.945
http://dx.doi.org/10.1037/0021-9010.93.5.945
http://dx.doi.org/10.1177/001872674900200204
http://dx.doi.org/10.1016/S0191-3085%2806%2927005-9
http://dx.doi.org/10.1016/S0191-3085%2806%2927005-9
http://dx.doi.org/10.2307/1556333
http://dx.doi.org/10.1016/j.socnet.2012.08.001
http://dx.doi.org/10.1016/j.socnet.2012.08.001
http://dx.doi.org/10.1016/S0378-8733%2802%2900019-9
http://dx.doi.org/10.1146/annurev.psych.56.091103.070250
http://dx.doi.org/10.1146/annurev.psych.56.091103.070250
http://dx.doi.org/10.1037/0021-9010.78.2.306
Johnson, D. W. (2003). Social interdependence:
Interrelationships among
theory, research, and practice. American Psychologist, 58, 934 –
945.

73. doi:10.1037/0003-066X.58.11.934
Johnson, D. W., & Johnson, R. T. (2005a). New developments
in social
interdependence theory. Genetic, Social, and General
Psychology
Monographs, 131, 285–358. doi:10.3200/MONO.131.4.285-358
Johnson, D. W., & Johnson, R. T. (2005b). Training for
cooperative group
work. In M. A. West, D. Tjosvold, & K. G. Smith (Eds.), The
essentials
of teamworking: International perspectives (pp. 131–147).
Chichester,
England: Wiley.
Johnson, M. D., Hollenbeck, J. R., Humphrey, S. E., Ilgen, D.
R., Jundt, D.,
& Meyer, C. J. (2006). Cutthroat cooperation: Asymmetrical
adaptation
to changes in team reward structures. Academy of Management
Journal,
49, 103–119. doi:10.5465/AMJ.2006.20785533
Jones, G. R. (1984). Task visibility, free riding, and shirking:
Explaining
the effect of structure and technology on employee behavior.
Academy
of Management Review, 9, 684 – 695.
Kane, G. C., & Labianca, G. (2005, October). Accounting for
clergy’s
social ledgers: Mixed blessings associated with direct and in
direct
negative ties in a religious organization. Paper presented at the
Intra-

74. Organizational Network (ION) Conference, Atlanta, Georgia.
Katz, R. (1982). The effects of group longevity on project
communication
and performance. Administrative Science Quarterly, 27, 81–104.
doi:
10.2307/2392547
Kozlowski, S. W. J., & Chao, G. T. (2012). The dynamics of
emergence:
Cognition and cohesion in work teams. Managerial and Decision
Eco-
nomics, 33, 335–354. doi:10.1002/mde.2552
Kozlowski, S. W. J., Gully, S. M., Nason, E. R., & Smith, E. M.
(1999).
Developing adaptive teams: A theory of compilation and
performance
across levels and time. In D. R. Ilgen & E. D. Pulakos (Eds.),
The
changing nature of performance: Implications for staffing,
motivation,
and development (pp. 240 –292). San Fransisco, CA: Jossey-
Bass.
Labianca, G., & Brass, D. J. (2006). Exploring the social ledger:
Negative
relationships and negative asymmetry in social networks in
organiza-
tions. Academy of Management Review, 31, 596 – 614.
doi:10.5465/
AMR.2006.21318920
Labianca, G., Brass, D. J., & Gray, B. (1998). Social networks
and
perceptions of intergroup conflict: The role of negative

75. relationships and
third parties. Academy of Management Journal, 41, 55– 67.
doi:10.2307/
256897
Lam, C. K., Van der Vegt, G. S., Walter, F., & Huang, X.
(2011). Harming
high performers: A social comparison perspective on
interpersonal
harming in work teams. Journal of Applied Psychology, 96, 588
– 601.
doi:10.1037/a0021882
LeBreton, J. M., & Senter, J. L. (2008). Answers to 20
questions about
interrater reliability and interrater agreement. Organizational
Research
Methods, 11, 815– 852. doi:10.1177/1094428106296642
LePine, J. A., Methot, J. R., Crawford, E. R., & Buckman, B.
(2012). A
model of positive relationships in teams: The role of
instrumental,
friendship, and multiplex social network ties. In L. T. Eby & T.
D. Allen
(Eds.), Personal relationships: The effect on employee attitudes,
behav-
ior, and well-being (pp. 173–194). New York, NY: Routledge.
Lyons, B. J., & Scott, B. A. (2012). Integrating social exchange
and
affective explanations for the receipt of help and harm: A social
network
approach. Organizational Behavior and Human Decision
Processes,
117, 66 –79. doi:10.1016/j.obhdp.2011.10.002

76. Major, D. A., Kozlowski, S. W. J., Chao, G. T., & Gardner, P.
D. (1995).
A longitudinal investigation of newcomer expectations, early
socializa-
tion outcomes, and the moderating effects of role development
factors.
Journal of Applied Psychology, 80, 418 – 431.
doi:10.1037/0021-9010
.80.3.418
Mathieu, J. E., & Taylor, S. R. (2006). Clarifying conditions
and decision
points for mediational type inferences in organizational
behavior. Jour-
nal of Organizational Behavior, 27, 1031–1056.
doi:10.1002/job.406
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS
procedures for
estimating indirect effects in simple mediation models.
Behavior Re-
search Methods, Instruments, & Computers, 36, 717–731.
doi:10.3758/
BF03206553
Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007).
Addressing
moderated mediation hypotheses: Theory, methods, and
prescrip-
tions. Multivariate Behavioral Research, 42, 185–227.
doi:10.1080/
00273170701341316
Reagans, R., & Zuckerman, E. W. (2001). Networks, diversity,
and pro-

77. ductivity: The social capital of corporate R&D teams.
Organization
Science, 12, 502–517. doi:10.1287/orsc.12.4.502.10637
Rousseau, V., & Aubé, C. (2010). Team self-managing
behaviors and
team effectiveness: The moderating effect of task routineness.
Group
& Organization Management, 35, 751–781. doi:10.1177/
1059601110390835
Saavedra, R., Earley, P. C., & Vandyne, L. (1993). Complex
interdepen-
dence in task-performing groups. Journal of Applied
Psychology, 78,
61–72. doi:10.1037/0021-9010.78.1.61
Seers, A. (1989). Team-member exchange quality: A new
construct for
role-making research. Organizational Behavior and Human
Decision
Processes, 43, 118 –135. doi:10.1016/0749-5978(89)90060-5
Seers, A., Petty, M. M., & Cashman, J. F. (1995). Team-member
exchange
under team and traditional management: A naturally-occurring
quasi-
experiment. Group & Organization Management, 20, 18 –38.
doi:
10.1177/1059601195201003
Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif,
C. W.
(1961). Intergroup conflict and cooperation: The Robbers Cave
exper-
iment. Norman: University of Oklahoma Book Exchange.

78. Sparrowe, R. T., Liden, R. C., Wayne, S. J., & Kraimer, M. L.
(2001).
Social networks and the performance of individuals and groups.
Acad-
emy of Management Journal, 44, 316 –325.
doi:10.2307/3069458
Totterdell, P., Hershcovis, M. S., Niven, K., Reich, T. C., &
Stride, C.
(2012). Can employees be emotionally drained by witnessing
unpleasant
interactions between coworkers? A diary study of induced
emotion
regulation. Work & Stress: An International Journal of Work,
Health &
Organisations, 26, 112–129.
doi:10.1080/02678373.2012.681153
Van der Vegt, G. S., Emans, B. J. M., & Van de Vliert, E.
(2001). Patterns
of interdependence in work teams: A two-level investigation of
the
relations with job and team satisfaction. Personnel Psychology,
54,
51– 69. doi:10.1111/j.1744-6570.2001.tb00085.x
Venkataramani, V., & Dalal, R. S. (2007). Who helps and harms
whom?
Relational antecedents of interpersonal helping and harming in
organi-
zations. Journal of Applied Psychology, 92, 952–966.
doi:10.1037/0021-
9010.92.4.952
Weiss, H. M., & Cropanzano, R. (1996). Affective events

79. theory: A
theoretical discussion of the structure, causes and consequences
of
affective experiences at work. Research in Organizational
Behavior, 18,
1–74.
Xia, L., Yuan, Y. C., & Gay, G. (2009). Exploring negative
group dynam-
ics adversarial network, personality, and performance in project
groups.
Management Communication Quarterly, 23, 32– 62.
doi:10.1177/
0893318909335416
Received December 21, 2012
Revision received January 20, 2014
Accepted January 21, 2014 �
T
hi
s
do
cu
m
en
t
is
co
py

80. ri
gh
te
d
by
th
e
A
m
er
ic
an
P
sy
ch
ol
og
ic
al
A
ss
oc
ia
ti

81. on
or
on
e
of
it
s
al
li
ed
pu
bl
is
he
rs
.
T
hi
s
ar
ti
cl
e
is

82. in
te
nd
ed
so
le
ly
fo
r
th
e
pe
rs
on
al
us
e
of
th
e
in
di
vi
du

83. al
us
er
an
d
is
no
t
to
be
di
ss
em
in
at
ed
br
oa
dl
y.
522 DE JONG, CURŞEU, AND LEENDERS
http://dx.doi.org/10.1037/0003-066X.58.11.934
http://dx.doi.org/10.3200/MONO.131.4.285-358
http://dx.doi.org/10.5465/AMJ.2006.20785533

84. http://dx.doi.org/10.2307/2392547
http://dx.doi.org/10.2307/2392547
http://dx.doi.org/10.1002/mde.2552
http://dx.doi.org/10.5465/AMR.2006.21318920
http://dx.doi.org/10.5465/AMR.2006.21318920
http://dx.doi.org/10.2307/256897
http://dx.doi.org/10.2307/256897
http://dx.doi.org/10.1037/a0021882
http://dx.doi.org/10.1177/1094428106296642
http://dx.doi.org/10.1016/j.obhdp.2011.10.002
http://dx.doi.org/10.1037/0021-9010.80.3.418
http://dx.doi.org/10.1037/0021-9010.80.3.418
http://dx.doi.org/10.1002/job.406
http://dx.doi.org/10.3758/BF03206553
http://dx.doi.org/10.3758/BF03206553
http://dx.doi.org/10.1080/00273170701341316
http://dx.doi.org/10.1080/00273170701341316
http://dx.doi.org/10.1287/orsc.12.4.502.10637
http://dx.doi.org/10.1177/1059601110390835
http://dx.doi.org/10.1177/1059601110390835
http://dx.doi.org/10.1037/0021-9010.78.1.61
http://dx.doi.org/10.1016/0749-5978%2889%2990060-5
http://dx.doi.org/10.1177/1059601195201003
http://dx.doi.org/10.1177/1059601195201003
http://dx.doi.org/10.2307/3069458
http://dx.doi.org/10.1080/02678373.2012.681153
http://dx.doi.org/10.1111/j.1744-6570.2001.tb00085.x
http://dx.doi.org/10.1037/0021-9010.92.4.952
http://dx.doi.org/10.1037/0021-9010.92.4.952
http://dx.doi.org/10.1177/0893318909335416
http://dx.doi.org/10.1177/0893318909335416When Do Bad
Apples Not Spoil the Barrel? Negative Relationships in Teams,
Team Performance, and ...HypothesesEffect of Negative
Relationships on Team PerformanceBuffering Negative
Relationships: Three ModeratorsNegative Relationships,
Cohesion, and Communication DensityNegative Relationships,

85. Cohesion, and Task-InterdependenceNegative Relationships,
Cohesion, and the Quality of Social ExchangeMethodProcedure
and ParticipantsMeasuresPresence of negative
relationshipsTeam cohesionCommunication densityTMXTask-
interdependenceTeam performanceControl
variablesAnalysesResultsDiscussionLimitations and Avenues
for Future ResearchConclusionsReferences