Journal of Retailing 80 (2004) 181–195

Managing conflict to improve the effectiveness of retail networks

Kevin D. Bradforda,∗ , Anne Stringfellowb,1 , Barton A. Weitzc,2
a Department of Marketing, 388 Mendoza College of Business, University of Notre Dame, Notre Dame, IN 46556, USA
b Global Business Department, Thunderbird, 15249 N. 59th Ave., Glendale, AZ 85306, USA
c Department of Marketing, University of Florida, 212 Bryan Hall, Gainesville, FL 32611, USA

Accepted 1 December 2003

Abstract

Retailers are becoming involved with networks consisting of multiple firms in order to more effectively perform business activities such
as supply chain management. This research develops and tests a framework outlining the effects of conflict in networks and how conflict
management can mitigate and exacerbate these effects. A study of 81 simulated networks finds that inter-personal and task conflict have a
negative effect on network member satisfaction and desire to be a member of the network. The use of a collaborative conflict management
style has a positive effect on satisfaction and desire for continuity, but the effects of accommodative and confrontational styles depend on the
level of inter-personal and task conflict present in the network.
© 2004 by New York University. Published by Elsevier. All rights reserved.

Keywords: Conflict, Conflict management, Retail network

Introduction composed of a focal retailer, the retailer’s vendors, logistics

suppliers, software and system providers, and organizations
Retailers increasingly are becoming involved with groups such as Voluntary Interindustry Commerce Standards (VICS)
of other firms to improve their effectiveness in performing that provide standards to facilitate communications [see the
business activities. This trend toward complex, interorgani- special issue on Supply Chain Management in the Journal
zational relationships suggests that firm performance may of Retailing, Levy & Grewal 2000]. However, retailers are
be determined by competition between business networks involved in other networks. For example, the effective use
rather than competition between individual firms (Achrol of electronic article surveillance (EAS) tags requires the
1997; Hakansson & Snehota 1995; Möller & Halinen 1999). coordinated efforts of the suppliers of tags and detection
A business network is defined as multiple independent equipment, the set of vendors affixing the tags to products,
businesses that have an interdependent relationship without and the retailers buying and selling the merchandise (Chain
hierarchical control (Anderson, Hakansson, and Johanson Store Age 1999; McNichols 2001). Other examples of a
1994). These networks are more than a set of firms involved in retail networks are retailers that share a common facility
exchanges. The firms in a business network have relationships such as tenants in a mall and groups of retailers that
that are characterized by cooperation as well as competition work together with manufacturers to develop private label
(McLoughlin & Horan 2002). merchandise for sales by all of the retailers in the buying
Perhaps the most widely discussed networks in retailing group.
involve supply chain management. These networks are The diverse capabilities and resources possessed by the
parties in these retail networks provide the potential for im-
proved performance. However, the firms in the network prob-
∗ Corresponding author. Tel.: +1 574 631 5057. ably will view the same issues through very different lenses.
E-mail addresses: [email protected] (K.D. Bradford),
[email protected] (A. Stringfellow), [email protected] (B.A. Weitz).
Their different perspectives can create conflict. Thus, conflict
1 Tel.: +1 602 978 7452. is a ubiquitous and important aspect of network dynamics
2 Tel.: +1 352 392 0161x1255. and the management of conflict plays an important role in

0022-4359/$ – see front matter © 2004 by New York University. Published by Elsevier. All rights reserved.
doi:10.1016/j.jretai.2003.12.002
182 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195

network performance (Achrol & Kotler 1999; Bengtsson & to work together in the future. These attitudes towards the
Kock 2000). network indicate the degree to which members will actively
The objectives of this research are to: (1) extend the re- support the network and continue to participate in the network
search on conflict in channel relationships by considering in the future.
effectiveness of conflict management styles in managing dif-
ferent types of conflict and (2) understand how the types of Types of conflict
conflict and conflict management affect network outcomes.
Building on previous research on conflict resolution (Rahim Conflict is defined as the behaviors or feelings of inter-
1986; Thomas 1976, 1992) and conflict in groups (De Dreu dependent parties in response to potential or actual obstruc-
1999; Jehn 1995, 1999; Pelled, Eisenhardt, & Xin 1999), in tions that impede one or more of the parties achieving their
the next section, we develop a framework describing different goals (Coughlan et al. 2001; Deutsch 1973; Gaski 1984)
types of conflict and the effects of conflict management ap- While most research in retailing and marketing on conflict
proaches on network outcomes. The hypotheses derived from focuses on channel relationships, conflict is a ubiquitous phe-
this framework are tested using a simulation of 81 networks. nomenon that pervades virtually all interorganizational ac-
After presenting the results, the paper concludes with a dis- tivities (Dahrendorf 1959; Lewin 1947; Thomas 1976). Even
cussion of the limitations and direction for future research. though most of the empirical research has centered on reduc-
ing the negative effects of conflict, conflict can also produce
positive outcomes by establishing an opportunity for express-
Proposed model and hypotheses ing grievance, critically reviewing past activities, and intro-
ducing different perspectives to produce innovative solutions
The conceptual framework on which this research is based (e.g., Coughlan et al. 2001, p. 241; Pondy 1967).
is presented in Fig. 1. The level of network conflict and the Conflict is a multi-dimensional construct (Amason,
conflict management approach used have a direct effect on Thompson, Hochwarter, & Harrison 1995; Jehn 1995). In
network outcomes. In addition, the conflict management ap- this research, we consider two dimensions, or types, of con-
proach used moderates the effect of conflict on network out- flict: (1) inter-personal conflict and (2) task conflict. These
comes. two dimensions parallel the classic factors affecting judg-
Considerable marketing research has examined the role of ments and decision-making—affect and cognitions. Thus,
conflict in channel relationships (see Geyskens, Steenkamp, inter-personal conflict is also called emotional or affective
& Kumar 1998 for a meta-analytic review) as well as intra- conflict, while task conflict is referred to as cognitive con-
organizational contexts such as new product development flict.
(e.g., Sethi 2000; Song, Xie, & Dyer 2000). However, this
prior research has simply viewed conflict as a unidimensional Inter-personal conflict
construct assessed by calculating an index based on the im- Inter-personal conflicts are disagreements within net-
portance, intensity, and frequency of disagreements on a list works based on non-task related incompatibilities. For
of issues (Coughlan, Anderson, Stern, & El-Ansary 2001, p. example, some parties in the network involved with design-
240). In addition, the extant marketing research has focused ing and implementing EAS tags might have negative feeling
on documenting the negative impact of conflict on economic toward some large retailers in the network because the retail-
and non-economic satisfaction, largely ignoring the use of ers have reputations of taking advantage of suppliers or they
conflict management to mitigate these negative effects and might have negative feeling towards an individual represent-
potentially produce positive results (see Dant and Schul 1992 ing a firm in the network because the individual is perceived
for an exception). of as unethical. Inter-personal conflicts typically produce
In our framework, we consider two dimensions of con- suspicion, distrust, and hostility among network members
flict – inter-personal conflict and task conflict – and three ap- (Brehmer 1976; Faulk 1982; Guetzkow & Gyr 1954). Re-
proaches for managing conflict—confrontation, accommo- search suggests that this type of conflict reduces the network’s
dation, and collaboration. The network outcomes examined ability to reach high-quality decisions and impedes the accep-
in this study are the satisfaction of network members with the tance of decisions among network members. Inter-personal
process and outcome and the degree to which the members conflict limits processing of new information, gives rise to
desire to continue the relationship in the future. hostile attributions concerning each other’s intentions and
behaviors reduces receptiveness to ideas advocated by other
Network outcomes network members who are disliked, decreases willingness
to tolerate opposition, and disturbs effective communication
The two network outcomes examined in this research are and cooperation within the network (e.g., Amason 1996;
network satisfaction and network continuity. Network satis- Amason & Schweiger 1994; Baron 1991; Eisenhardt &
faction is defined as the extent to which network members Bourgeois 1988; Jehn 1995). Thus, we hypothesize that:
feel the processes and outcomes of interactions with other
network members exceed their expectations. Network con- H1. Inter-personal conflict in networks negatively affects
tinuity is the degree to which network members are willing the (a) network satisfaction and (b) network continuity.
K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195 183

Fig. 1. Conceptual framework.

184 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195

Task conflict Collaboration

Task conflict is defined as disagreements over how to ac- Collaboration conflict management reflects activities in
complish work together (e.g., Amason 1996). It often arises which network members explore integrative solutions. When
from systemic or structural incompatibilities (Wilkof, Brown, networks use this style of conflict management, they try to
& Selsky 1995). Network members often have different or- find new and creative solutions to problems by focusing on
ganizational and societal cultures, resources, and capabilities their needs as well as on the needs of the entire network
that create differences in perspectives on how to accomplish (Gross & Guerrero 2000). This exploration of win-win solu-
work. For example, retailers in the EAS tag network may not tions tends to maintain continuity in relationships (Hocker &
understand the difficulties vendors face in designing tags that Wilmot 1998). This style of managing conflict has the great-
will work with different detection systems. Another example est potential to produce positive outcomes. Its use can lead to
of task conflict is a disagreement within a network in a re- integrative solutions that can be mutually beneficial for the
tail mall setting over the nature of entertainment provided entire network. Thus, the following hypothesis:
to attract different consumer segments. Addressing this type
of conflict is effortful and can distract attention from per- H3. Collaboration positively affects (a) network satisfaction
forming the network’s objectives, thus negatively affecting and (b) network continuity.
network performance and satisfaction. Therefore, we hypoth-
esize that: While we hypothesize that using collaboration conflict
management is effective in all situations, accommodation and
H2. Task conflict in networks negatively affects (a) network confrontation conflict management, discussed in the follow-
satisfaction and (b) network continuity. ing sections, have a positive effect in some situations and a
negative effect in others.
Conflict management and network outcome
Accommodation
As hypothesized previously, conflict left unmanaged leads Accommodative conflict management behavior is defined
to negative consequences for the parties involved. However, as the behaviors that network members use to create an en-
research suggests that conflict, if managed properly, can have vironment where each party allows others to have their way
positive as well as negative consequences. For example, con- and/or accept the other members’ perspectives. The use of ac-
flict prevents stagnation, stimulates interest and curiosity in commodation conflict management can be effective because
a task, and provides a medium through which problems can it creates a context in which parties perceive that others are
be aired and solutions arrived at (Coser 1956; Deutsch 1973; willing to listen, accept their points of view, and make con-
Jehn 1997; Simmel 1955; Tjosvold 1985). Positive outcomes cessions to move forward on issues (Papa & Pood 1988).
of conflict also include an expanded understanding of the We propose that the use of accommodation conflict manage-
issues confronting the network, mobilization of network ment behaviors is beneficial to the relationship when network
resources and energies toward problem resolution, and clar- members experience inter-personal conflict. Inter-personal
ification of competing solutions and creative searches for al- conflict, by definition, is not concerned with the work to
ternative solutions to existing problems (Brown 1983). Thus, be accomplished and thus devoting time to resolving inter-
it is important for networks to manage conflict effectively personal conflicts distract attention from accomplishing the
so that the positive consequences of conflict can be realized network’s tasks. Focusing attention on inter-personal conflict
(Rahim 1986; Robbins 1978; Tjosvold 1985). For example, affects network members’ desire to communicate or interact
Schweiger et al. (1986, 1989) found that the management with one other and this can stagnate and sometimes destroy
of conflict, specifically the use of interaction techniques that the business relationships in the network.
force network members to disagree and debate the merits Using accommodative conflict management behaviors
of different alternative perspectives, produces superior signals a willingness to accept the perspectives of others
results. Thus, conflict management practices can mitigate (Hocker & Wilmot 1998). When inter-personal conflict oc-
the negative effects of conflict on network performance and curs, the process of making concessions, associated with
can result in exploiting the beneficial aspects of conflict. accommodation conflict management, reduces the negative
emotion in the network that impedes the attainment of net-
Conflict management behaviors work outcomes (Gross & Guerrero 2000). Thus, network
members will be able to concentrate on task-related network
A number of taxonomies of conflict management behav- issues. By reducing the tension caused by non-task issues,
iors have been developed (Blake & Mouton 1964; Hall 1969; network member satisfaction with outcomes and processes
Pruitt 1983; Rahim & Bonoma 1979; Stern, Sternthal, & as well as the desire for network continuity improves (Gross
Craig 1973; Thomas 1976, 1979). In this research, we ex- & Guerrero 2000).
amine the effects of three conflict management approaches
discussed in these frameworks: (1) collaboration, (2) con- H4. The use of accommodation conflict management mod-
frontation, and (3) accommodation. erates the impact of inter-personal conflict on (a) network
 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195 185

member satisfaction and (b) desire for continuity. The mod- found that communication strategies that emphasize con-
erating effect of accommodation conflict management is frontational behaviors are key to the development of trust
positive—greater use of accommodation conflict manage- and commitment (Helper & Sako 1995; Ping 1993, 1997).
ment reduces the negative impact of inter-personal conflict Network relationships that are open to opposing viewpoints
on satisfaction and desire for continuity. and that have members who are willing to communicate di-
rectly about the status of the network and their perspectives
On the other hand, the use of accommodation conflict are the basis for high quality interaction and the development
management behaviors to address task conflicts can be of sound relationships.
counter-productive and lead to network member dissat- However, if the level of task conflict is low and the level of
isfaction. Accommodative conflict management behavior inter-personal conflict is low, the use of confrontational con-
may inhibit the free exchange of information about the flict management will have a negative effect on network out-
importance of issues to specific network members. The comes. Confrontation, in this situation, creates controversy
degree to which task conflicts can be managed to obtain when significant controversy does not exist. Therefore, even
integrative solutions is lessened because of this reluctance to though network members have low task and inter-personal
identify each party’s perspectives on the critical issues and conflict, by expressing difference of opinion on how to ac-
approaches for performing the task. Without these different complish the task, the level of task conflict is raised need-
insights on how the task should be accomplished, the lessly. This leads to less satisfaction with network outcomes
development of mutually beneficial, integrative outcomes is and processes and a decreased desire to work together in the
reduced, thus leading to lower satisfaction and a lower desire future.
for network continuity (Gross & Guerrero 2000). Thus, Thus we propose that:

H5. The use of accommodation conflict management mod- H6. The use of confrontational conflict management mod-
erates the impact of task conflict on (a) network member sat- erates the impact of task conflict on (a) network member
isfaction and (b) desire for continuity. The moderating effect satisfaction and (b) desire for continuity, when inter-personal
of accommodation conflict management is negative—greater conflict is low and task conflict is high. The moderating effect
use of accommodation conflict management increases the of confrontational conflict management is positive—greater
negative impact of task conflict on satisfaction and desire for use of confrontational conflict management in this situation
continuity. decreases the negative impact of task conflict on satisfaction
and desire for continuity.
Confrontation
Confrontation conflict management is defined as the be- H7. The use of confrontational conflict management mod-
haviors that network members use in order to emphasize the erates the impact of task conflict on (a) network member
perspective of one or more of the network members. Thus, satisfaction and (b) desire for continuity when inter-personal
this style involves “the clashing of ideas” [dictionary defini- conflict is low and task conflict is low. The moderating effect
tion], not the unrelenting advocacy of a perspective. While we of confrontational conflict management is negative—greater
propose that task conflict, in general, has a negative effect on use of confrontational conflict management in this situation
network outcomes, confrontation conflict management has increases the negative impact of task conflict on satisfaction
been found to be effective in managing task conflicts in some and desire for continuity.
situations (Papa & Canary 1995). Research shows that when
parties engage in confrontational behavior and do so in an In the following sections, we describe the method used to
environment in which the parties are relaxed, friendly, and at- test the hypotheses, the results, their limitations, directions
tentive (i.e., low inter-personal conflict), confrontational be- for future research and managerial implications.
havior can yield positive outcomes (Infante & Gorden 1985).
The low level of inter-personal conflict, the parties are more
willing to express and discuss difference of opinion on how Method
to accomplish the task without an emotional reaction to dis-
agreements. This section describes the research design, sample and
The use of confrontation conflict management when task data collection procedure, the measure development, the mea-
conflict is high and inter-personal conflict is low results in sures used in the research, and the data analysis approach
“creative abrasion” (Leonard & Rayport 1997), the clash of used.
ideas that produces creative, synergistic outcomes. In addi-
tion, this willingness to reveal perspectives and concerns al- Research design
lows network members to better understand other network
members’ agendas. Further, confronting with respect to tasks In light of the limited research involving conflict in net-
associated with accomplishing work together creates a cli- works, we used a simulation to test the hypotheses proposed
mate of openness between network members. Research has in the previously presented conceptual framework. By using
186 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195

a simulation, we were able to measure conflict management them individually. The four network members knew their own
behaviors used in networks in close time proximity to the net- payoffs for each decision, but not those of the other network
work interactions and assess measures of network outcomes. members.
To test the hypotheses, we created 81 four-person net- The network exercise began with the participants receiv-
works. The networks, composed of MBA students, engaged ing instructions for the simulation, information about their
in a mixed-motive exercise developed by Beggs, Brett, and role, and the utilities associated with each level of the de-
Weingart (2000). Each member was assigned a different role cision variables. The participants then met with the other
within a network, and represented a different organization. members of their network and reached an agreement con-
Each also has the objective of optimizing the position of cerning the five decision variables. Finally, each participant
their organization within the network while, at the same time, completed a questionnaire to assess the constructs in the con-
maintaining the benefits of network membership. As in actual ceptual framework.
networks, the organizations interact through individual rep- This simulated network environment is consistent with the
resentatives who act as boundary-spanners operating at the definition of business networks in the literature. The multiple
interorganizational level. The network members were moti- entities are independent but interdependent without hierar-
vated to consider both the network’s goal and their individual chical control and, through cooperation, they can expand the
firm perspectives by awarding prizes to all the members of total utility of the group. In addition the simulation mirrored
the network with the best overall network performance and to the approach taken by actual networks by designating rep-
the best performance by the network member in each of the resenting of their organizations to meet together to address
four roles. The exercise placed each network member in the issues. Even though networks are composed of organizations,
role of a different specialty storeowner in a hypothetical retail issues are addressed by representatives of the organization,
building (mall). The mall would have an open plan with mer- not by organizations.
chants located in different “departments” around a common
area with tables for customers to sit down and sample some Measures
of the items sold in the specialty stores. Since there would
be several stores in the same location, customers would be Measures were developed using a framework outlined in
likely to purchase from more than one of the merchants. The Churchill (1979). First, a pool of items was developed for
four member stores stand to gain from the arrangement, not each construct. Past marketing literature on conflict (e.g.,
only from increased sales, but also from reduced building Barclay 1991; Clare & Sanford 1984; Crittenden et al. 1993)
maintenance costs. proved helpful in developing items to measure conflict itself,
The four storeowners had to discuss and come to agree- and we consulted the marketing, psychology, and manage-
ment on five issues that differentially affected each of their ment literatures (e.g., Amason 1996a; Rahim 1983; Ruekert
businesses. The five issues were the extent to which (1) adver- & Walker 1987) to develop items measuring conflict behav-
tising, (2) hiring and training of clerks, and (3) maintenance iors. These items were pre-tested on a convenience sample of
would be organized separately by each storeowner or jointly business school faculty and graduate students. Exploratory
for the mall as a whole, (4) the temperature within the mall, factor analyses and an examination of the item intercorrela-
and (5) the location of stores within the mall. For each owner, tions, means, and standard deviations were used to purify the
a value or utility was assigned for each outcome; however, the scales. Scale unidimensionality was verified using confirma-
values differed across owners. For example, both the flower tory factor analysis (Gerbing & Anderson 1988).
shop and coffee shop owners’ utility values were higher for The measure used for each network was the average score
locations near the entrance of the mall because the increased on each item across all participants in the network. The scales
traffic would generate more impulse visits. On the other hand, used to measure the constructs and reliabilities of the scales
the liquor store owner’s utility was higher for locations near are reported in the appendix. The means, standard deviations,
the loading dock since its merchandise (cases of alcoholic and correlation matrix for the constructs are shown in Table 1.
beverages) are difficult to move around and the store is a des- The items used to measure the constructs are shown in the
tination with limited impulse visits. The trade-off facing the Appendix.
flower shop and the coffee shop owners over location is an
example of one of the zero-sum situation; however, within Types of conflict
the payoff matrices, there were some situations in which two The five items used to measure inter-personal conflict as-
or three members of the network would all have increasing sessed the degree to which interorganizational friction and
utility from an agreement on an issue. An example of this is tension caused by interorganizational non-task incompatibil-
the hiring and training of clerks. The flower shop clerks are ities or dislikes occurred during the completion of the network
required to build elegant floral designs. This requires special- task. The Cronbach alpha for the scale is .94. The four-item
ized skills and experience, so the flower store is motivated scale measuring task conflict assessed the extent to which
to hire, train, and pay its clerks individually. In contrast, the members of the network had differences of opinions con-
other stores are not as concerned with clerks’ skills and would cerning the processes for accomplishing work together. The
prefer to hire and pay all the mall clerks as a group, but train responses for the items in these scales were collected on seven
 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195 187

Table 1
Means, standard deviations, and correlation matrix
Mean SD IPC TC AM CF CL SOP
Conflict type
Inter-personal conflict (IPC) 2.32 .94
Task conflict (TC) 4.21 .79 .61
Conflict management behavior
Accommodative (AM) 5.03 .51 −.19 −.28
Confrontational (CF) 3.68 .75 .59 .56 −.25
Collaborative (CL) 5.62 .60 −.46 −.34 .58 −.44
Outcome
Satisfaction with outcome and process (SOP) 4.90 .76 −.60 −.59 .38 −.44 .62
Network continuity (NC) 5.83 .64 −.54 −.41 .38 −.27 .52 .68

point scales anchored by “strongly disagree” and “strongly seven-point scale anchored by “strongly disagree” and
agree.” The Cronbach alpha for the task conflict scale is .81. “strongly agree” (Cronbach alpha .84).

Conflict management behaviors Formation of network measures

The three conflict management behaviors were measured
on seven-point scales assessing the degree to which the con- The responses obtained from multiple individual network
flict management style was used in the interaction. The scales members were to be aggregated into a single network-level
were anchored by “never” and “very frequently.” The seven response; therefore, to ensure that we were working with a
items used to measure accommodation conflict management construct at the network level, rather than the network mem-
behavior assessed the frequency of behaviors related to con- ber level, it was necessary to verify the relative magnitudes
forming to the views of others. The Cronbach alpha for this of the between-network and within-network variance. Fol-
scale is .85. Confrontational conflict management behavior lowing the procedure recommended by Georgopolous (1986)
(Cronbach alpha .83) was measured using a six-item scale to and used by Jehn (1995), a one-way ANOVA analysis was
assess the frequency of behaviors in which one or more net- performed for each network member measure using the mea-
work members asserted themselves in a conflict situation in sure as the dependent variable and the network as the sin-
order that their point of view might prevail. Finally, collabo- gle factor. The F-ratio from the ANOVA was compared with
ration conflict management behavior was measured using by Georgopolous’s criterion that F exceeds 1.0. In addition, the
five items (Cronbach alpha .90) assessing the frequency with eta-squared statistic was calculated for each measure. Eta-
which network members attempted to develop integrative so- squared, or the ratio of the between-network variance to the
lutions. total variance, should exceed 0.16 if data aggregation is to be
appropriate (Georgopolous 1986). All the measures have F
Network outcomes ratios that exceed 1.0 and eta-squared greater than .16.
The two dependent measures used in this research Hypotheses H1 through H5 were tested by estimating the
were satisfaction and network performance. The degree to following multiple regression models:

Satisfaction = b00 + b01 × inter-personal conflict (IPC) + b02 × task conflict (TC) + b03 × accommodation (AM)
+ b04 × confrontation (CF) + b05 × collaboration (CL) + b06 × IPC × AM + b07 × IPC × CF
+ b08 × IPC × CL + b09 × TC × AM + b010 × TC × CF + b011 × TC × CL (1)

Network continuity = b10 + b11 × inter-personal conflict (IPC) + b12 × task conflict (TC) + b13 × accommodation (AM)
+ b14 × confrontation (CF) + b15 × collaboration (CL) + b16 × IPC × AM + b17 × IPC × CF
+ b18 × IPC × CL + b19 × TC × AM + b110 × TC × CF + b111 × TC × CL (2)

where b00 b111 represent the unstandardized regression coef-

which network members were satisfied with the process ficients for the respective independent variables.
and outcome was measured using a six-item, seven-point The measures of the individual constructs were mean
scale anchored by “strongly disagree” and “strongly agree” centered to reduce multicollinearity as recommended by
(Cronbach alpha .89). Finally, network continuity, the degree Cronbach (1987) and Jaccard, Turrisi, and Wan (1990).
to which the network members are interested in working Hypotheses H6 and H7 involve third order interactions—
together in the future, was assessed using a three-item, interaction between confrontation conflict management,
188 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195

inter-personal conflict, and task conflict. When the third- satisfaction (β01 = −0.383, p < .005) and has a significant neg-
order interactions were added to Eqs. (1) and (2), the ative effect on desire for network continuity (β11 = −0.428,
multicollinearity was so great that none of the estimated p < .005). Task conflict is significantly related to network
coefficients was significant. Thus, to test these hypotheses, satisfaction (β02 = −0.332, p < .005), but is not significantly
we used the medians of the types of conflict to divide related to network continuity (β12 = −0.138).
the sample into four groups: (1) high task conflict—low
relationship conflict, (2) low task conflict—high relationship
Main effects of conflict management behaviors on
conflict, (3) low task conflict—high relationship conflict,
network outcomes
and (4) low task conflict—low relationship conflict. Then we
estimated the correlation between the conflict management
The results concerning the main effect of collaboration
styles and network outcomes for each group. The results of
conflict management on network outcomes partially support
these analyses are shown in Tables 3 and 4.
H3. The use of collaboration conflict management is
significantly related to network member satisfaction (β05 =
0.304, p < .01). While the estimated effect of collaboration
Results on network continuity is in the hypothesized direction
(β15 = 0.244), it is not statistically significant. While we
The standardized coefficients estimated for Eqs. (1) and did not hypothesize a main effect for accommodation and
(2) concerning the factors affecting network outcomes cov- confrontation conflict management on network outcomes,
ered in the first five hypotheses are shown in Table 2. confrontation conflict management had a significant positive
effect on network continuity (β14 = 0.252, p <.05).
Main effects of conflict on network outcomes
Moderating effects of conflict management behaviors
H1 and H2, the negative main effects of conflict types
on network outcomes, are largely supported. Inter-personal
The results support H4 and H5 that the use of accom-
conflict has a significant negative effect on network member
modation conflict management moderates the effects of task
and inter-personal conflict on network outcomes. Accommo-
Table 2 dation significantly and positively moderates the effects of
Network outcomesa
inter-personal conflict on network satisfaction (β06 = 0.261,
Dependent variable p < .05) and desire for network continuity (β16 = −0.293, p <
Satisfaction with Network .05). As hypothesized, accommodation conflict management
process and continuity significantly and negatively moderates the effects of task con-
outcome flict on network satisfaction (β09 = 0.262, p < .05) and desire
Inter-personal conflict (IPC) −.383∗∗∗ −.428∗∗∗ for network continuity (β19 = 0.266, p < .05).
Task conflict (TC) −.332∗∗∗ −.138 While we did not hypothesize any moderating effects for
Accommodation (AM) .056 .197 collaboration, collaboration did significantly and negatively
Confrontation (CF) .093 .252∗ moderates the relationship between inter-personal conflict
Collaboration(CL) .304∗∗ .244
and network continuity (β111 = −0.374, p < .05). This re-
IPC × AM .261∗ .293∗ sult suggests that, while the use of collaboration is generally
IPC × CF .224 −.098 a useful conflict management approach (H3), its effectiveness
IPC × CL −.186 −.374∗
is diminished with higher levels of inter-personal conflict.
TC × AM −.262∗ −.266∗ As mentioned previously, hypotheses H6 and H7 involve
TC × CF −.209 −.002
third order interactions since they predict that the moderating
TC × CL .080 −.228
effects of confrontation conflict management will depend on
R2 .645 .491 the levels of inter-personal and task conflict. The results of the
Adjusted R2 .588 .410
Significance of F for model .000 .000
correlations for the four different conflict situations shown in
a The maximum variance inflation factor (VIF) for the set of independent
Tables 3 and 4 support H6 and H7. In networks with high task
variables used in these regression models was 4.075. Since this value is below
conflict and low inter-personal conflict, confrontation conflict
10, multicollinearity should not have an undue influence on the least squares management is positively related to network satisfaction (r
estimates (Neter, Wasserman, & Kutner 1990). However, when three-way = .58, p < .05) and desire for network continuity (r = .71, p
interaction terms were added to the regression analysis, the maximum vari- < .01). However, in networks with low task conflict and low
ance inflation factor increased above the accepted limit of 10, indicating inter-personal conflict, confrontation has a negative effect on
that multicollinearity was influencing the least squares estimates. We thus
resorted to testing hypotheses H6 and H7 by means of correlation analysis.
network satisfaction (r = −.47, p < .01) and network conti-
∗ p < .05. nuity (r = −.30, p < .10). Of the three conflict management
∗∗ p < .01. behaviors shown in the tables, only confrontation shows this
∗∗∗ p < .005.
sign reversal.
 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195 189

Table 3
Correlation between conflict management styles and network satisfaction

Discussion The misuse of conflict management approaches can

exacerbate the negative effects of network conflict. For
The results of the network simulations support the concep- example, the network outcomes are diminished when
tual framework we presented. Inter-personal and task conflict accommodation conflict management is used to address task
have a negative effect on network outcomes. These negative conflict. Confrontation conflict management has negative
effects can be reduced by the use of the appropriate conflict effects on network outcomes when used under condition of
management approaches. The use of collaboration conflict low inter-personal and low task conflict.
management is effective in reducing both types of conflict. The results also emphasize the need to consider the
However, accommodation conflict management is effective multi-dimensional nature of conflict. Task and inter-personal
only at reducing inter-personal conflict and confrontation is conflict, the two dimensions of network conflict consid-
effective only at reducing task conflict in networks where ered in this research, need to be approached differently.
inter-personal conflict is low and task conflict is high. While both of these types of conflict have a negative ef-
190 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195

Table 4
Correlation between conflict management styles and network continuity

fect on network outcomes, conflict management approaches resulting in weak member performance. Similarly, the main
are differentially effective in mitigating or exacerbating these effects of task conflict on network outcomes might be
effects. greater for networks in which there is truly an opportunity
to integrate different capabilities to develop innovative
Limitations strategies.
The single-period nature of the simulation presents limi-
While the results support our conceptual framework, the tations for the generalization of findings to ongoing network
framework was tested in only one context—a mixed-motive relationships. In our study, there was no opportunity to under-
exercise involving a simulation of networks composed of stand how the actions in one time period may affect a future
MBA students. One would expect that the specific results time period, which is the essence of any kind of ongoing ex-
would be different when networks have a different compo- change or network relationship. A multi-period design would
sition and are embedded in different contexts. For example, allow researchers to better simulate real-world network envi-
the impact of inter-personal conflict would probably be ronments in which networks might benefit from knowledge
greater for networks facing considerable competition, gained in previous periods (Hunt 1995). In addition, the range
 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195 191

of alternatives was restricted. Networks members could not ent perspectives on how to implement the system create task
exercise their creativity and develop new and potentially in- conflict.
tegrative solutions. At first, the vendors reduce the level of task conflict by
Finally, the outcomes examined in the research are sub- accommodating and accepting the position of the EAS tag
jective assessments of satisfaction and desire for continuity. suggested by the EAS detection equipment manufacturers.
We did not directly examine the objective performance of the However, as the meeting progresses the vendors see their
networks. By assessing subjective performance, there is the costs mounting and the retailers unwilling to pay for the ad-
potential for common method variance biasing the results to- ditional costs. They bring up the issue of the placement of
ward significance. While this might affect the strength of the the tags again. The level of task conflict increases as the EAS
main effect relationships, it is unlikely to affect the strength manufacturers and vendors dispute each others claims. Both
of the hypothesized interactions. parties use a confrontational approach to managing the in-
creasing task conflict by providing more support for their
Managerial implications positions. The information conveyed as the parties support
their positions give rise to the parties collaboratively explor-
As indicated previously, major packaged good manufac- ing different locations for the tags and uncovering a loca-
turers (i.e., Proctor & Gamble, Eastman Kodak, and John- tion that is both low cost and does not result in deactivation
son & Johnson), major retailers (i.e., Wal-Mart, Walgreens) problems. In this situation, accommodation was not effec-
and manufacturers of EAS tag detection systems are in- tive in managing task conflict. Confrontation was effective
volved in a network cost reduction initiative. In order to effec- in this situation because of the low level of inter-personal
tively implement this new technology they must coordinate conflict. Through confrontation a collaborative solution
their efforts so as to maximize the potential gains for the arose.
project (Chain Store Age 1999). The following hypotheti- This scenario illustrates the following conceptual and em-
cal scenario concerning representatives of the firms involved pirical results of this paper:
implementing EAS tags illustrates the implications of this
• Parties in networks need to recognize that conflicts, both
research.
task and non-task (inter-personal), will arise and that these
Assume representatives from the retail firms, the vendors,
conflicts are detrimental to network outcomes (satisfaction
and the manufacturers of EAS detection systems meet to dis-
within the network and desire to continue the effort).
cuss the issues related to implementing EAS tags. As the
• Active management of the conflict by network members
meeting progresses, a vendor’s representative claims that one
can reduce the level of the conflict and even increase sat-
of the retailers is unethical because it engages in excessive
isfaction and desire for continuity.
charge backs. The vendor raises concerns about whether the
• Use collaborative conflict management behaviors when
retailer will ever be fair in allocating the costs and benefits
facing both task and non-task conflict.
of putting EAS tags on merchandise. The retailer’s repre-
• Only use accommodative conflict management behaviors
sentative vendor is incensed by this accusation and calls the
in the presence of high inter-personal conflict and only
vendor representative a liar. In a previous meeting, the same
use confrontational conflict management behaviors in the
discussion arose and as each party confronted each other
presence of low inter-personal related conflict.
with information supporting their positions, the level of inter-
• Avoid using confrontation when task and non-task related
personal conflict escalated to the point that the vendor in-
conflict is low.
volved dropped out of the network. However, in this meeting,
this behavior resulting in a high level of inter-personal con- Managing inter-organizational network relationships is
flict is diffused when another vendor in the meeting uses an a particular challenge, since their very nature includes
accommodating approach to managing the conflict. The other conflicting objectives in addition to a need for coopera-
vendor suggests the conflicting parties accept that charge tion. Adding to this difficulty, the lack of a hierarchical
backs are part of all vendor–retailer relationships and says control structure means that network members are them-
“Vendors don’t like charge backs and retailers feel they need selves tasked with managing network relationships. Since
to use charge backs to insure complete shipments. So let’s marketers have considerable expertise in managing inter-
just agree to disagree and get on with the task at hand.” Af- organizational relationships, they are well positioned to play
ter accommodating the inter-personal conflict, the network a major role in the network management activities of their
members get down to the task of implementing the EAS tag firms.
system. Conflict is multi-dimensional construct. The types or di-
The vendors indicate they need to put the tags at a par- mensions of conflict can differentially affect the quality of
ticular place on their merchandise to reduce costs. However, the relationship between network members. Thus, knowing
the EAS tag manufacturers claim that placing the tags in that the type of conflict that is present in network relationships
spot on the merchandise will result in a higher failure rate is important to managers because of its effects their use
to deactivate the tags when the merchandise is bought and a of conflict management approaches and subsequent network
larger number of false positives for shoplifting. These differ- outcomes.
192 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195

Most firms train their people to work toward collabora- team members get better the potential for conflict increases
tive relationships and to collaborate effectively. This research (Braiker & Kelley 1979; Coser 1956; Jehn & Shah 1996).
shows that collaboration, though ideal in many situations, is These concepts may also apply to network conflict and may
not the only conflict management style that produces positive have implications on network performance.
outcomes. Firms should train their representatives when it is
most appropriate to use confrontation and accommodation as
Creativity as a network outcome
well as collaboration.
The creativity with which a business activity is undertaken
is an important, subjective measure of network performance.
Directions for future research
Creativity is defined as the creation of novel and useful out-
comes (Amabile 1983). In fact, one could argue that the pri-
While the results of this research support the basic
mary reason for bringing together the different abilities and
premises of the conceptual framework, the framework cer-
resources represented in a business network is to facilitate
tainly needs to be tested in an actual network environment.
the development of creative solutions.
The simulation with MBA students used in this study has
The optimum conditions for creativity include the pres-
the advantages of enabling a straightforward data collec-
ence of a non-judgmental atmosphere (Amabile 1996, 1998;
tion process and the collection of conflict management be-
Delbecq & Mills 1985), which encourages people to present
haviors shortly after the completion of the task. However,
their unedited thoughts without fear of ridicule. In situ-
the results from using a simulation may be biased because
ations where disagreements about network goals exist, it
a simulation does not capture the same level of involve-
is likely that people will be less willing to expose them-
ment that is present in actual network activities. On the
selves to criticism, and therefore we propose that networks
other hand, the results are promising and suggest that addi-
with high levels of task conflict will tend to exhibit low
tional research in the area of network conflict management is
creativity.
merited.
Three other issues that merit further investigation are: (1)
factors causing task and inter-personal conflict, (2) other net- Nature of the task undertaken
work outcomes such as the creativity of the task solution, and The framework could be extended to consider mod-
(3) the impact of the nature of the tasks undertaken by the erating effects of the nature of the activities in which
network. the network is involved. Hambrick, Davidson, Snell, and
Snow (1998) divide group tasks into coordinative, compu-
Antecedents of network conflict tational, and creative tasks. The activity considered in this
Network members contribute different skills and perspec- research is a computational task with limited opportuni-
tives to the performance of business activities. Thus, network ties for truly innovative solutions. The level and nature of
members differ from one another in important respects. Some conflict might change and the effective management of the
of these differences may impact network conflict. Research conflict might differ as the opportunity for innovative so-
on the effects of team composition on team conflict can be lutions increases. The moderating effects of these different
used to develop hypotheses about the link between network task types, and solution possibilities, on the nature of con-
composition and network conflict. flict and conflict management in networks need to be inves-
Recent research on teams comprised of diverse members tigated.
has recognized the importance of “unobservable diversity” The present model does not take into account the trend to-
(Milliken & Martins 1996) and “deep level diversity” ward globalization, which means that the members of many
(Harrison, Price, & Bell 1998) amongst team members. inter-organizational networks are located remotely from one
This literature uses the term “creative abrasion” (Leonard & another, and may have different cultural norms for relation-
Rayport 1997) to describe the conflicts arise from collisions ships and conflict management. The lack of opportunity for
of disparate team-member perspectives. face-to-face communication, combined with different expec-
In addition to differences in perspectives amongst net- tations, may increase the challenge of managing conflict in
work members, differences in power and commitment are these networks. This is an important area requiring future
other potential sources of task conflict in the network. For research.
example, networks with high power differences may also Effective conflict management in networks is a complex
experience more inter-personal friction than networks com- phenomenon meriting additional research attention. This re-
posed of equal status members (Dutton & Walton 1966; search presents a framework for advancing the understand-
McCann & Galbraith 1981; Smith, Carroll, & Ashford ing of conflict management strategies and some preliminary
1995). Another example elucidating antecedents leading to results supporting the framework. Further empirical inves-
different perspectives is that the more committed members tigations and conceptual extensions of the framework can
may propose more complex approaches to the task, or more potentially improve the theoretical understanding of the role
ambitious goals, leading to task conflict. An often over- and management of conflict in networks, thereby enhancing
looked antecedent to conflict is that as relationship between their productivity.
 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195 193

Appendix A. Measurement scales

Scale items Cronbach ␣

Inter-personal conflict .94

Our personalities occasionally clashed.
There was a lot of tension between us at times.
At times, there were bad feelings between us.
Sometimes we became irritated with one another.
There was personal friction among network members.
Task conflict .81
We disagreed over ideas.
Group members disagreed over how to complete the project.
Group members had differences of opinion over how to complete the project.
We experienced differences of opinion.
Accommodative conflict management behavior .85
Giving in to others.
Accepting others’ positions.
Letting the other person win.
Agreeing to go along with the others’ ideas.
Conforming to others’ wishes.
Complying with others’ ideas.
Accommodating the wishes of the rest of the group.
Confrontational conflict management behavior .83
Trying to get one’s own way.
Arguing a point without considering the views of others.
Pushing one’s own interests regardless of the effect on other group members.
Pitting one’s own viewpoint against those of other group members.
Trying to win arguments.
Confronting other group members with dissenting views.
Collaborative conflict management behavior .90
Collaborating.
Reaching agreement by pooling our ideas together.
Integrating our ideas to come up with a jointly acceptable decision.
Coming up with a creative solution that we all can agree on.
Actively seeking a mutually beneficial solution.
Satisfaction with outcome and process .89
I was disappointed with the way the project turned out (reverse-scored).
I was very satisfied with the outcome of this project.
We could have done better on this project (reverse-scored).
I was very satisfied with the process used to complete the project.
We could have used a more effective method to accomplish our task (reverse-scored).
I would have preferred the project to have been done differently (reverse-scored).
Network continuity .84
I would be willing to work with this group on future projects.
It would be good to get the same group together to work on future projects.
I do not want to work with this group again.
194 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195

References Crittenden, Victoria L., Gardiner, Lorraine R., & Stam, Antonie. (1993).
Reducing conflict between marketing and manufacturing. Industrial
Achrol, Ravi. (1997). Changes in the theory of interorganizational re- Marketing Management, 22(4), 209–299.
lations in marketing: Toward a network paradigm. Journal of the Cronbach, Lee J. (1987). Statistical tests for moderator variables: Flaws in
Academy of Marketing Science, 25(Winter), 6–71. analyses recently proposed. Psychological Bulletin, 102(3), 414–417.
Achrol, Ravi, & Philip, Kotler. (1999). Marketing in the network econ- Dahrendorf, Ralf. (1959). Class and class conflict in industrial society.
omy. Journal of Marketing, 63(Special Issue), 146–165. Stanford: Stanford University Press.
Amabile, Teresa M. (1983). The Social Psychology of Creativity: A com- Dant, Rajiv, & Schul, Patrick. (1992). Conflict resolution processes in con-
ponential conceptualization. Journal of Personality & Social Psychol- tractual channels of distribution. Journal of Marketing, 56(January),
ogy, 45(August), 357–376. 38–54.
Amabile, Teresa M. (1996). Assessing the work environment for creativ- De Dreu, Carsten K. W. (1999). Productive conflict: The importance of
ity. Academy of Management Journal, 39(October), 1154–1185. conflict management and conflict issue. In Carsten K. W. De Dreu &
Amabile, Teresa M. (1998). How to kill creativity. Harvard Business Evert Van de Vliert (Eds.), Using conflict in organizations (pp. 9–22).
Review, (September–October), 76–87. London: Sage.
Amason, Allen C. (1996a). Distinguishing the effects of functional and Delbecq, Andre L., & Mills, Peter K. (1985). Managerial practices that
dysfunctional conflict on strategic decision making: Resolving a para- enhance innovation. Organizational Dynamics, 14(1), 24–34.
dox for top management networks. Academy of Management Journal, Deutsch, Morton. (1973). The resolution of conflict. New Haven, CT: Yale
39(February), 123–148. University Press.
Amason, Allen C. (1996b). Discussing the effects of conflict on strategic Dutton, John M., & Walton, Richard E. (1966). Interdepartmental conflict
decision making performance. In Carsten De Dreu & Evert Van de and cooperation: Two contrasting studies. Human Organization, 25,
Vliert (Eds.), Optimizing performance by stimulating conflict: Fun- 207–220.
damental issues for groups and organizations. CA, Sage: Thousand Eisenhardt, Kathleen M., & Bourgeois, L. J., III. (1988). Policies of deci-
Oaks. sion making in high velocity environments. Academy of Management
Amason, Allen C., & Schweiger, David M. (1994). Resolving the paradox Journal, 31, 737–750.
of conflict, strategic decision making, and organizational performance. Faulk, Richard E. (1982). The effects of a two-weekend marital enrich-
International Journal of Conflict Management, 5(July), 239–253. ment program on self-disclosure and marital adjustment. Dissertation
Amason, Allen C., Thompson, Kenneth R., Hochwarter, Wayne A., & Abstracts International, 42(8–13), 3398–3399.
Harrison, Allison W. (1995). Conflict: An important dimension in Gaski, John F. (1984). The theory of power and conflict in channels of
successful management networks. Organizational Dynamics, 24(2), distribution. Journal of Marketing, 48(Summer), 9–29.
20–35. Georgopolous, Basil S. (1986). Organizational structure, problem solving,
Anderson, James C., Hakansson, Hakan, & Johanson, Jan. (1994). Dyadic and effectiveness. San Francisco, CA: Jossey-Bass.
business relationships within a business network context. Journal of Gerbing, David W., & Anderson, James C. (1988). An updated paradigm
Marketing, 58(October), 1–15. for scale development incorporating unidimensionality and its assess-
Barclay, Donald W. (1991). Interdepartmental conflict in organizational ment. Journal of Marketing Research, 25(May), 186–192.
buying: The impact of the organizational context. Journal of Market- Geyskens, Inge, Steenkamp, Jan-Benedict E. M., & Kumar, Nirmalya.
ing Research, 28(May), 145–159. (1998). Generalizations about trust in marketing channel relationships
Baron, Robert A. (1991). Positive effects of conflict. Employee Respon- using meta analysis. International Journal of Research in Marketing,
sibilities and Rights Journal, 4(1), 25–36. 15(1), 223–248.
Beggs, Rebecca, Brett, Jeanne M., & Weingart, Laurie. (2000). Towers Gross, Michael, & Guerrero, Laura K. (2000). Managing conflict appro-
market. In Negotiation & decision making exercises 2000 Version. priately and effectively: An application of the competence model to
Chicago, IL: Northwestern University Dispute Resolution Center. Rahim’s organizational conflict styles. International Journal of Con-
Bengtsson, Maria, & Kock, Soren. (2000). ‘Coopetition’ in business flict Management, 11(3), 200–227.
networks—Cooperate and compete simultaneously. Industrial Mar- Guetzkow, Harold, & Gyr, John. (1954). An analysis of conflict in
keting Management, 29, 411–426. decision-making groups. Human Relations, 7, 367–382.
Blake, Richard R., & Mouton, Jane S. (1964). The managerial grid. Hakansson, Hakan, & Snehota, Ivan. (1995). Developing relationships in
Houston, TX: Gulf. business networks. New York, NY: Routledge.
Braiker, H. B., & Kelley, H. H. (1979). Conflict in the development of Hall, Jay. (1969). Conflict management survey: A survey of one’s char-
close relationships. In R. L. Burgess & T. L. Huston (Eds.), Social acteristic reaction to and handling of conflicts between himself and
exchange in developing relationships (pp. 135–168). New York, NY: others. Houston, TX: Teleometrics.
Academic Press. Hambrick, Donald C., Davidson, Sue C., Snell, Scott A., & Snow, Charles
Brehmer, Berndt. (1976). Social judgment theory and the analysis of in- C. (1998). When groups consist of multiple nationalities: Towards a
terpersonal conflict. Psychological Bulletin, 83(November), 985–1003. new understanding of the implications. Organization Studies, 19(2),
Brown, David L. (1983). Managing conflict at organizational interfaces. 181–205.
Reading, MA: Addison-Wesley Publishing Company. Harrison, David A., Price, Kenneth H., & Bell, Myrtle P. (1998). Beyond
Chain Store Age. (1999, May). Suppliers Seek EAS Standard (pp. relational demography: Time and the effects of surface- and deep-level
160–161). diversity on work group cohesion. Academy of Management Journal,
Churchill, Gilbert A. (1979). A paradigm for developing better measures 41(1), 96–107.
of marketing constructs. Journal of Marketing Research, 16(February), Helper, Susan R., & Sako, Mari. (1995). Supplier relations in Japan and
64–73. the United States: Are they converging. Sloan Management Review,
Clare, Donald A., & Sanford, Donald G. (1984). Cooperation and conflict 36(3), 77–85.
between industrial sales and production. Industrial Marketing Man- Hocker, Joyce L., & Wilmot, William W. (1998). Interpersonal conflict
agement, 13, 163–169. (5th ed.). Madison, WI: Brown and Benchmark.
Coser, Lewis A. (1956). The functions of social conflict. New York, NY: Hunt, Kenneth A. (1995). The relationship between channel conflict and
Free Press. information processing. Journal of Retailing, 71(4), 417–436.
Coughlan, Anne, Anderson, Erin, Stern, Louis, & El-Ansary, Adel I. Infante, Dominic A., & Gorden, William I. (1985). Superiors’ argumen-
(2001). Marketing channels (6th ed.). Upper Saddle River, NJ: Pren- tativeness and verbal aggressiveness as predictors of subordinates’
tice Hall. satisfaction. Human Communication Research, 12, 117–125.
 K.D. Bradford et al. / Journal of Retailing 80 (2004) 181–195 195

Jaccard, James, Turrisi, Robert, & Wan, Choi K. (1990). Interaction ef- Pondy, Louis R. (1967). Organizational conflict: Concepts and models.
fects in multiple regression. Newbury Park, CA: Sage. Administrative Science Quarterly, 12(2), 296–320.
Jehn, Karen A. (1995). A multimethod examination of the benefits and Pruitt, Dean G. (1983). Strategic choice in negotiation. American Behav-
detriments of intragroup conflict. Administrative Science Quarterly, ioral Scientist, 27(2), 167–194.
40(June), 256–282. Rahim, M. Afzalur. (1986). Managing conflict in organizations. New
Jehn, Karen A. (1997). A qualitative analysis of conflict types and di- York, NY: Praeger Publishers.
mensions in organizational groups. Administrative Science Quarterly, Rahim, M. Afzalur. (1983). Measurement of organizational conflict. Jour-
42(3), 530–557. nal of General Psychology, 109(October), 189–199.
Jehn, Karen A. (1999). Why differences make a difference: A field study Rahim, M. Afzalur, & Bonoma, Thomas V. (1979). Managing organiza-
of diversity, conflict, and performance in workgroups. Administrative tional conflict: A model for diagnosis and intervention. Psychological
Science Quarterly, 44(4), 741–763. Reports, 44, 1325–1344.
Jehn, Karen A., & Shah, Priti Pradhan. (1996). Interpersonal relation- Robbins, Stephen P. (1978). Conflict management and conflict resolution
ships and task performance: an examination of mediating processes are not synonymous terms. California Management Review, 21(2),
in friendship and acquaintance groups. Academy of Management Pro- 67–75.
ceedings, 279–284. Ruekert, Robert W., & Walker, Orville C., Jr. (1987). Marketing’s in-
Leonard, Dorothy, & Rayport, Jeffrey F. (1997). Spark innovation through teraction with other functional units: A conceptual framework and
empathic design. Harvard Business Review, 75(6), 102–114. empirical evidence. Journal of Marketing, 51(1), 1–19.
Lewin, Kurt. (1947). Frontiers in group dynamics. Human Relations, 1(1), Schweiger, David M., Sandberg, William R., & Rechner, Paula L. (1989).
1–45. Experiential effects of dialectical inquiry, devil’s advocacy, and con-
Levy, Michael, & Grewal, Dhruv. (2000). Supply chain management in sensus approaches to strategic decision making. Academy of Manage-
a networked economy. Journal of Retailing, 76, 415–430. ment Journal, 32, 745–772.
McCann, Joseph, & Galbraith, Jay R. (1981). Interdepartmental relations. Schweiger, David M., Sandberg, William R., & Ragan, James W. (1986).
In Paul C. Nystrom & William H. Starbuck (Eds.), Handbook of Group approaches for important decision making: A comparative anal-
organizational design (pp. 60–84). Oxford: Oxford University Press. ysis of dialectical inquiry, devil’s advocacy, and consensus. Academy
McLoughlin, Daniel, & Horan, Conor. (2002). Markets as networks: Notes of Management Journal, 29(1), 51–71.
on a unique understanding. Journal of Business Research, 55(July), Sethi, Rajesh. (2000). Superordinate identity in cross-functional product
535–543. development teams: Its antecedents and effect on new product perfor-
McNichols, Shawn. (2001). Store of the Future, Security, 35–36 (Decem- mance. Journal of the Academy of Marketing Science, 28(3), 330–344.
ber). Simmel, George. (1955). Conflict. New York, NY: Free Press.
Milliken, Frances J., & Martins, Luis L. (1996). Searching for common Smith, Ken G., Carroll, Stephen J., & Ashford, Susan A. (1995).
threads: Understanding the multiple effects of diversity in organiza- Intra- and interorganizational cooperation: Toward a research agenda.
tional groups. Academy of Management Review, 21(2), 402–433. Academy of Management Journal, 38(February), 7–23.
Möller, Kristian E., & Halinen, A. (1999). Business relationships and Song, X. Michael, Xie, Jinhong, & Dyer, Barbara. (2000). Antecedents
networks: Managerial challenge of network era. Industrial Marketing and consequences of marketing managers’ conflict-handling behaviors.
Management, 28(September), 413–427. Journal of Marketing, 64(1), 50–67.
Neter, John, Wasserman, William, & Kutner, Michael H. (1990). Applied Stern, Louis W., Sternthal, Brian, & Craig, C. Samuel. (1973). Manag-
linear statistical models. Burr Ridge, IL: Irwin. ing conflict in distribution channel: A laboratory study. Journal of
Papa, Michael J., & Canary, D. J. (1995). Conflict in organizations: A Marketing Research, 10(May), 169–179.
competence-based approach. In A. M. Nicotera (Ed.), Conflict and Thomas, Kenneth W. (1976). Conflict and conflict management. In Mar-
organizations: communicative processes (pp. 153–179). Albany, NY: vin D. Dunnette (Ed.), Handbook of industrial and organizational
State University of New York Press. psychology (pp. 889–935). Chicago, IL: Rand McNally.
Papa, Michael J., & Pood, Elliot A. (1988). Coorientational accuracy Thomas, Kenneth W. (1979). Conflict. In Kerr Steven (Ed.), Organiza-
and differentiation in the management of conflict. Communication tional behavior (pp. 151–181). Columbus, OH: Grid Publications.
Research, 15(August), 400–426. Thomas, Kenneth W. (1992). Conflict and negotiation processes in orga-
Pelled, Lisa H., Eisenhardt, Kathleen M., & Xin, Katherine R. (1999). nizations. In Marvin D. Dunnette & Leaetta M. Hough (Eds.), Hand-
Exploring the black box: An analysis of work group diversity, conflict, book of industrial and organizational psychology: Vol. 3 (2nd ed., pp.
and performance. Administrative Science Quarterly, 44(1), 1–28. 651–717). Palo Alto, CA: Consulting Psychologists Press.
Ping, Robert A. (1993). The effects of satisfaction and structural con- Tjosvold, Dean. (1985). Implications of controversy research for manage-
straints on retailer exiting, voice, loyalty, opportunism, and neglect. ment. Journal of Management, 11, 19–35.
Journal of Retailing, 69(Fall), 320–353. Wilkof, Marsha, Brown, Deborah, & Selsky, John. (1995). When the
Ping, Robert A. (1997). Voice in business-to-business relationships: cost- stories are different: The influence of corporate culture mismatches on
of-exit and demographic antecedents. Journal of Retailing, 73(2), inter-organizational relations. Journal of Applied Behavioral Science,
261–281. 31(September), 373–388.