See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/260787388
Coordination in Consultant-Assisted IS Projects: An Agency Theory
Perspective
Article in IEEE Transactions on Engineering Management · May 2010
CITATIONS
READS
8
129
2 authors:
Matthew J. Liberatore
Wenhong Luo
115 PUBLICATIONS 2,034 CITATIONS
30 PUBLICATIONS 765 CITATIONS
Villanova University
SEE PROFILE
Villanova University
SEE PROFILE
All content following this page was uploaded by Matthew J. Liberatore on 14 March 2014.
The user has requested enhancement of the downloaded file.
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 57, NO. 2, MAY 2010
255
Coordination in Consultant-Assisted IS Projects:
An Agency Theory Perspective
Matthew J. Liberatore and Wenhong Luo
Abstract—Increasingly, consulting firms are employed by client
organizations to participate in the implementation of enterprise
systems projects. Such consultant-assisted information systems
projects differ from internal and outsourced IS projects in two important respects. First, the joint project team consists of members
from client and consulting organizations that may have conflicting
goals and incompatible work practices. Second, close collaboration between the client and consulting organizations is required
throughout the course of the project. Consequently, coordination
is more complex for consultant-assisted projects and is critical for
project success. Drawing from coordination and agency theories
and the trust literature, we developed a research model to investigate how interorganizational coordination could help build relationships based on trust and goal congruence and achieve higher
project performance. Hypotheses derived from the model were
tested using data collected from 324 projects. The results provide
strong support for the model. Interorganizational coordination
was found to have the largest overall significant effect on performance. However, its effect was achieved indirectly by building trust
and goal congruence and by reducing technical and requirements
uncertainty. The positive effects of trust and goal congruence on
project performance demonstrate the importance of managing the
client–consultant relationship in such projects. Project uncertainty,
including both technical and requirements uncertainty, was found
to negatively affect goal congruence and trust, as expected. This
study represents a step toward the development of a new theory on
the role of interorganizational coordination.
Index Terms—Agency theory, consulting, enterprise systems, interorganizational coordination, management of information systems projects, trust.
I. INTRODUCTION
NCREASINGLY, the function of information systems (IS)
implementation is shifting from building customized systems to implementing packaged enterprise systems, such as
SAP [70]. This change has had a significant impact not only
on the software development process [10], [69], [70] but also
on the implementation process within an organization [9], [34].
Due to the increased complexity and scale of enterprise systems,
a key factor for a successful implementation is the utilization
of third-party consulting firms to provide the necessary technical expertise and project management skills [9]. As a result,
I
Manuscript received April 18, 2008; revised August 29, 2008. First published
March 27, 2009; current version published April 21, 2010. Review of this
manuscript was arranged by Department Editor J. K. Pinto.
M. J. Liberatore is with the Department of Management and Operations/International Business, Villanova University, Villanova, PA 19085 USA
(e-mail: matthew.liberatore@villanova.edu).
W. Luo is with the Department of Accounting and Information Systems, Villanova University, Villanova, PA 19085 USA (e-mail: wenhong.luo@
villanova.edu).
Digital Object Identifier 10.1109/TEM.2009.2013838
consulting firms are playing a growing and critical role in IS
development and implementations.
According to the Gartner group, U.S. companies spent about
$45 billion on information technology (IT) consulting services
in 2004, and expenditures are expected to grow to $58 billion by 2009 [28]. Consulting firms provide technical expertise and/or project management skills. In this research, we
refer to IS implementation projects that involve significant
and close collaboration between client and consulting organizations as consultant-assisted IS projects. Consultant-assisted
IS projects differ from internal and outsourced IS projects
in two important respects. First, the joint project team consists of members from client and consulting organizations that
may have conflicting goals and incompatible work practices.
Second, close collaboration between the client and consulting organizations is required throughout the course of the
project.
Managing client–consultant relationships during implementation has proven to be a very complex and difficult task, because the parties may have differing goals, expectations, work
processes, and prior experiences with IS development [30], [50].
In several well-publicized cases, companies have attributed
their implementation failures to their consulting partners’ poor
advice, inexperienced personnel, and misaligned incentives.
FoxMeyer, for example, sued Anderson Consulting and Deloitte
after a failed SAP implementation that led to the company’s
bankruptcy [72]. Consequently, effectively managing client–
consultant relationships is critical for implementation success.
Coordination can play a critical role in building and maintaining
good relationships between client and consultants and improving project performance.
The effect of coordination has been studied for internal IS
projects (e.g., [1], [60], and [61]). These studies focused on the
intraorganizational coordination between the business users and
IS staff. In consultant-assisted projects, coordination also needs
to occur at the interorganizational level between the consultants
and client users and IS staff. Such coordination is more complex,
because consultants and clients work for different organizations
and may have different objectives and work under different
incentive mechanisms.
Client and consulting organizations participating in
consultant-assisted IS projects have a principal–agent relationship, indicating that goals of the agent may conflict with those
of the principal and the actions of the agent can be difficult for
the principal to observe. This study attempted to answer the
following question:
Does interorganizational coordination lead to improved performance
of consultant-assisted IS projects by addressing the agency problem
0018-9391/$26.00 © 2009 IEEE
256
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 57, NO. 2, MAY 2010
by helping to build trust and goal congruence between the parties,
and by reducing the effect of project uncertainty?
In the following sections, we first provide a review of the coordination, agency theory, and trust literatures in the IS project
domain. Then, we advance a research model and hypotheses
that conceptualize the effects of interorganizational coordination on the project performance. The research and data analysis
methods for model testing are described in detail. The study
results as well as implications for managers and researchers are
discussed. Finally, limitations of the current study and future
research directions are presented.
II. THEORETICAL BACKGROUND
In the following sections, we first review studies of the effects
of coordination in internal and outsourced IS projects. Next, we
discuss agency theory and its applications to IS projects and
then discuss the role of trust between project team members and
how it can affect project performance.
A. Coordination Theory
Coordination has been a subject of study in many disciplines,
such as economics, organizational theory, computer science,
and artificial intelligence [52]. In organizational theory, coordination is often conceptualized as the management of interdependence among different activities to accomplish certain
objectives [27], [52], [80]. In the project management literature, Berggren et al. [6] discuss how the increasing use of
consultants leads to problems of project coordination that cannot be addressed using only comprehensive contracts and plans.
Chiocchio [12] found that high-performing project teams started
to coordinate themselves later than low-performing teams but
maintained higher levels of coordination afterward.
Given that IS projects involve cross-functional teams, several studies have examined the effects of team-level coordination on project performance. However, the conceptualization of
coordination and the factors that could mediate and moderate
the relationship between coordination and performance differed
across these studies. Specifically, while some studies have focused on the techniques used to coordinate project activities,
others examined how the level of coordination impacts project
performance. Kraut and Streeter [44] defined coordination techniques in terms of activities that assist members of software
development teams to interact and communicate. These techniques were grouped into five categories: formal impersonal
procedures, formal interpersonal procedures, informal interpersonal procedures, electronic communication, and interpersonal
network. They found that the use of coordination techniques
was associated with the stages and attributes of the project. For
example, formal procedures and interpersonal networks were
used in larger projects, while interpersonal procedures were
employed during the planning stage of the project. In addition, interpersonal networks were used more often when the
project was highly uncertain and involved a lot of interdependent
tasks.
Using the structural contingency and risk-based software engineering perspectives, Nidumolu [60], [61] showed that the
effects of coordination on project performance could be directed or mediated by project uncertainty and project risk.
While different coordination mechanisms might impact various factors, a certain level of coordination was required in
the same project to achieve high project performance. Parolia et al. [62] also studied the relationship between coordination and IS project performance. They found that coordination enhanced the level of leadership empowerment and
knowledge transfer and helped to clarify the mission and objectives among team members. These factors, in turn, improved project performance, which was not directly impacted by
coordination.
Andres and Zmud [1] defined coordination strategies along
three dimensions: the degree of centralization, formality of the
communication channels, and the degree of control in decision making. In a laboratory experiment, they examined the
relationships between task interdependence, goal conflict, and
coordination strategies. Their study showed that there was an
interaction effect between coordination that involved close collaboration and task interdependence, i.e., close coordination was
especially effective when the projects involved highly interdependent tasks. The explanation was that coordination enabled
team members to better pool available knowledge and skills and
reduced the effects of complexity associated with interdependent tasks. Both Nidumolu [60], [61] and Andres and Zmud [1]
found that coordination had positive impact on the performance
of internal IS projects, while Parolia et al. [62] did not find a
direct effect.
Jha and Iyer [38] argued that coordination among project
participants has a considerable effect on the outcome of a
construction project. They identified the set of coordination
activities that contribute the most to the overall coordination rating for a project. In another study, Jha and Iyer [39]
also found that coordination was a key factor in achieving
the schedule, cost, and quality objectives of a construction
project.
The research on interorganizational coordination in the IS
project environment is limited. Sabherwal [68] examined the
coordination mechanisms used in outsourced IS projects based
on 11 case studies. He found that clients and vendors had rather
different perspectives on what are the appropriate coordination mechanisms and when they should be used. For example,
clients seemed to prefer informal way of coordination with the
vendors. On the other hand, vendors stressed the importance of
standards and plans in order to formally establish the expectation
with respect to the system and project management. Mirani [57]
studied coordination relating to the management of day-to-day
cooperative and collaborative activities using two case studies of
off-shored software tasks. He found the critical, enabling role of
coordination in facilitating successful interchanges between onshore and offshore teams for software-related tasks. This study
showed the importance of carefully specifying and partitioning
tasks, and using integrating mechanisms, such as having the
vendor interface directly with onshore client, to bridge communication gaps.
LIBERATORE AND LUO: COORDINATION IN CONSULTANT-ASSISTED IS PROJECTS: AN AGENCY THEORY PERSPECTIVE
B. Agency Theory
Agency theory is concerned with addressing the “agency
problem,” in which the goals of the agent conflict with those
of the principal and the actions of the agent are difficult for
the principal to observe [22], [46]. Information asymmetries
and goal incompatibility are identified as the two key issues in
the agency theory [73]. Information asymmetries suggest that
principals know far less about agents’ capabilities and actions
than agents themselves do. Thus, when an agent has incompatible goals with the principal, the agent may take advantage
of the information asymmetries and pursue opportunistic behavior, such as “shirking.” Agency theory has been widely
applied in the management literature to examine and explain
shareholder–manager, Board of Directors–CEO, headquarters–
subsidiary, and other principal–agent relationships [22], [73].
Classic agency theory was developed from the economics
perspective and is based on a number of assumptions that critics
have long argued are too simplistic and narrow in defining principal and agent [21], [73]. First, it is assumed that both principals
and agents are rational individuals who will always maximize
their self-interests. It discounts the possibility that some individuals may behave altruistically under certain circumstances.
Second, the classic agency theory focuses on the dyadic relationship between a single principal and an agent. But, of course, the
roles of principal and agent can be rather complicated and may
be difficult to define. Despite these limitations, agency theory is
still a useful lens through which we can explain the behavior of
individuals in many organizational settings.
In the project management literature, agency theory has been
applied to explain the relationships between project owners and
managers (e.g., [59] and [79]) and a potential agency risk problem in project finance [25]. In the systems development literature, agency theory has been applied to explain the relationships between project managers and developers in internal IS
projects [2], [51], and between clients and vendors in outsourced
IS projects [45], [55], [68].
The agency problem in internal IS projects is manifested
by the goal conflicts between project managers and developers. Such goal conflicts may arise when developers are given
multiple goals, or when project managers and developers have
incompatible goals. For example, project managers want to ensure that the IS project is delivered on time and on budget in
order to meet their performance criteria. Developers, however,
may be more concerned about creating high-quality systems and
maintaining their marketability. Such differences in goals, if not
managed, can lead to time and budget overruns [51].
Goal conflicts can occur between vendors and clients in outsourced IS projects as well [45], [55]. As in internal IS projects,
clients are interested in developing high-quality systems on time
and within budget [17], [63]. Vendors, on the other hand, are
also motivated by profits and client satisfaction [30]. Due to their
lack of expertise, clients often rely on vendors for technical and
project management skills. As a result, clients may feel particularly vulnerable to vendors’ opportunistic behaviors [68], and so
vendor governance becomes a key success factor in outsourced
IS projects [13].
257
Agency theory suggests a number of ways to address the
agency problem, including outcome-based contracts, monitoring, and trust [11], [22], [41], [78]. In outcome-based contracts,
the agent’s compensation is linked to the principal’s goals and
objectives. As such, the agent has a clear incentive to act in the
principal’s best interest [22]. However, outcome-based contracts
are more difficult to execute in IS projects, because systems requirements are often not known in advance and may change
over time. Mahaney and Lederer [51] found that although an
array of supplemental incentives such as technical training and
flexible work schedules were used for motivation, the primary
incentive mechanism for internal IS developers was behaviorbased straight salaries. McFarlan and Nolan [55] suggested that
it is important for the client to have flexibility built into outsourcing contracts, because the desired outcomes might change
due to evolving technology and business conditions. Therefore,
outcome-based contracts alone are not sufficient to resolve the
agency problem in IS projects.
The principal can also monitor the work of the agent by itself or through a third party [41], [58], [78]. Monitoring deters
the agent from pursuing opportunistic behavior and provides
a method for identifying problems and solutions quickly. It is
especially effective in conjunction with an outcome-based contract [78]. Kirsch et al. [42] examined formal and informal
control modes used in IS development projects. They found
that the selection of control mode depends on behavior observability, outcome measurability, and client’s understanding
of the IS development process. Mahaney and Lederer [51] reported that monitoring was widely used by project managers
to track progress and supervise developers. For outsourced IS
projects, Sabherwal [68] showed that the client used coordination as a way to track vendor behavior, leading to appropriate
adjustments. He also observed that the client might change the
coordination mechanisms in response to performance problems
or perceived opportunistic behavior.
C. Trust
While controlling outcomes and behavior has been extensively studied in the agency theory literature, building trust is an
alternative approach to address the agency problem [11]. Trust
is “the willingness of a party to be vulnerable to the actions of
another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of
the ability to monitor or control that other party” [53]. Trust is
a necessary condition for the principal and agent to form meaningful economic relationships [66]. As a result, the desire to
gain trust can provide the intrinsic motivation for the agent to
perform in the principal’s best interest, and thereby preventing
the agent from pursuing opportunistic behaviors [11].
Lewicki et al. [49] review and organize the trust literature.
They review the Lewicki and Bunker [47], [48] transformation model of trust development that consists of calculus-based,
knowledge-based, and identification-based trust. Calculusbased trust calculates the outcomes resulting from creating and
sustaining a relationship related to the costs of maintaining or
severing it. Knowledge-based trust knows the other sufficiently
258
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 57, NO. 2, MAY 2010
well so that other’s behavior is predictable. Identification-based
trust identifies with the other’s desires and initiates mutual understanding so that one can act for the other.
Lewicki et al. [49] also address the work of Rousseau et
al. [67], who also considered calculus-based trust as part of their
approach. However, Rousseau et al. [67] argue that calculusbased trust is formed on rational choice: trust emerges when
the trustor perceives that the trustee tends to perform an action
that is beneficial to the trustor. Here, individuals are motivated
by economic self-interest and can be based on economic incentives or contractual sanctions. The second component of their
approach is relational trust that arises between individuals who
interact repeatedly over time. Personal experience and information forms the basis for relational trust, and emotions and
personal attachments influence the relationship. The third component is institution-based trust that results from the efforts of
institutions to shape the conditions needed for trust to develop.
The legal system, societal norms relating to conflict management and cooperation, and systems regulating education and
the professions support the development of institution-based
trust.
In a study of blended project teams including the client and IT
employees, Weber [81] found that client trust in his or her project
manager is positively related to project team performance. Following the work of McAllister [54], trust was measured along
two dimensions: affective or relational trust used by Rousseau
et al. [67] and cognitive trust, which is grounded in individual beliefs about peer reliability and dependability, as well as
competence.
Also based on a review of the trust literature, Mayer et al. [53]
suggested that trust is influenced by perceived trustworthiness
that includes three dimensions: ability, benevolence, and integrity (see also [35] and [56]). Ability refers to the set of skills
and competencies that would enable the agent to make good
on promises. In order to gain the principal’s trust, the agent is
motivated to demonstrate that it possesses the requisite skills
and abilities. Benevolence is the extent to which the agent is
perceived to have good intentions toward the principal without
extrinsic reward (e.g., a profit motive). Benevolence can be the
result of the agent’s desire to establish or maintain a long-term
mutually rewarding relationship with the principal. An agent is
said to have integrity when the agent is believed to adhere to a
set of principles or ethical standards that are acceptable to the
principal. The agent with integrity is intrinsically motivated to
provide the appropriate level of effort desired by the principal.
Conceptualizing trust as based on the three dimensions of
Mayer et al. [53], Gefen [30] explored the relationship between
trust and engagement success in ERP implementations from the
client’s perspective. He showed that consulting firms can influence the client’s perception of engagement success by gaining
the client’s trust, which, in turn, was affected by the responsiveness and dependability of the services provided. The results of
his study suggested that it would be in the best interest of the
consulting firms to gain trust from the client to ensure future
business and long-term success.
Zaghloul and Hartman [85] and Hartman [33] identified three
bases of trust that might apply to partners (principal and agent)
in the construction industry: competence, intuitive (emotional),
and integrity (ethical) trust. Competence trust is based on the
perception of the partner’s ability to perform the required work,
while intuitive trust is founded upon the party’s prejudice, biases, and personal feelings toward their counterparts, developed
over a long-term relationship. Integrity trust is based on the
perception of the partners’ willingness to protect the interest of
their counterpart during the execution of the project. Each of
these is quite similar to the corresponding dimensions in the
Mayer et al. [53] framework.
Kadefors [40] argued that contractual incentives and close
monitoring of contractor performance in construction projects
may signal a climate of distrust that tends to induce opportunism
and hinder cooperative interaction, and that a higher level of trust
would improve project performance. Partnering practices such
as team-building and project-wide communication in the early
project phases were found to have potential in influencing the
antecedents of trust and creative teamwork. Based on a review
of the various trust attributes found in the construction project
literature, Wong and Cheung [83] determined the relative importance of trust factors for clients and consultants, as well as
contractors engaged in construction projects. Clients and consultants ranked system-based trust as most important, indicating
that they rely on satisfactory contract terms to enhance trust.
For contractors, system-based trust is also highest ranked, while
partners’ performance and permeability (openness, information
flow) is a close second.
In summary, the coordination, agency theory, and trust literatures have focused on internal and outsourced IS projects, with
very limited research on consultant-assisted projects. In addition, the existing studies have taken either a coordination, agency
theory, or trust perspective. Given that the client and consulting
organizations participating in consultant-assisted projects have
essentially a principal–agent relationship, we theorize that coordination can alleviate the agency problem by promoting goal
congruence and trust between the parties.
III. RESEARCH MODEL AND HYPOTHESES
Based on coordination and agency theories, we have developed a research model (Fig. 1) that examines how higher levels
of coordination can lead to higher project performance through
its effects on the client–consultant agency relationship by building trust and goal congruence. This model also suggests that
interorganizational coordination can reduce the effect of technical and requirements uncertainty on project performance. In
addition, technical and requirements uncertainty are thought to
negatively impact the trust and goal congruence between the
client and consultant.
As discussed by Gable [26], prior research has suggested the
importance of distinguishing between the results of a consulting engagement and the effectiveness of, or satisfaction with,
the consultant’s performance in arriving at those results. Other
research has suggested distinguishing between the project and
the outcomes. Doll and Torkzadeh [20] define project success
as “being able to complete projects that meet their design requirements on time and within budget.” For these reasons,
we measure project performance in term of process quality,
LIBERATORE AND LUO: COORDINATION IN CONSULTANT-ASSISTED IS PROJECTS: AN AGENCY THEORY PERSPECTIVE
Fig. 1.
259
Rresearch model.
product quality, and the extent to which the project budget and
schedule are met. Process quality reflects the quality of the
client–consultant interactions, while product quality refers to
the quality of the implemented IS applications and satisfaction
of business users [14], [60].
Interorganizational coordination refers to activities and techniques used by client and consultant participants to manage
their interdependence. In studying internal software development projects, Nidumolu [60] showed that higher levels of coordination led to higher levels of project performance. Similarly, for the purpose of this study, we focus on the levels of
coordination between client and consultant rather than specific
coordination techniques and methods. Like internal IS projects,
consultant-assisted projects rely on multiple participants working together effectively to ensure project success. However, the
consultant-assisted project environment is more complex than
that of internal IS projects, because project participants belong
to different organizations that may have differing goals and work
processes. This organizational complexity suggests that interorganizational coordination would be critical to project success.
H1: Higher levels of interorganizational coordination will lead
to higher levels of project performance.
Agency theory suggests that goal conflicts can be the sources
of shirking and moral hazard in a principal–agent relationship [22]. As noted earlier, the objectives of the client and
consulting organizations are typically different from each other.
In general, the client wants to obtain a high-quality system at
the lowest possible cost, while the consultant desires profits and
valuable experiences [17], [30], [63]. Goal congruence refers
to the extent to which the client and consulting organizations
perceive the possibility of having compatible, if not identical,
goals with respect to the project [23], [37], [71]. Goal congruence is often the product of ongoing interactions through which
both parties can understand each other’s objectives and identify opportunities for aligning compatible and complementary
objectives [36], [37].
The level of client–consultant coordination can impact the
degree of goal congruence in two major ways. First, coordination provides the necessary mechanisms and communication
channels to foster interactions between the client and consultant
organizations throughout the project. The more coordination
takes place within the project, the more likely that clients and
consultants can understand the goals and priorities of the other
side and identify common grounds for goal alignment. Second, coordination can serve as effective monitoring measures
for uncovering and managing goal conflicts [51]. Frequent coordination ensures that any goal conflicts are discovered early
before they develop into serious problems that could negatively
impact project performance.
H2: Higher levels of interorganizational coordination will lead
to higher levels of goal congruence.
Building trust has been identified as a useful approach to
address the agency problem [11]. Since trust is a necessary condition for meaningful and lasting economic relationships between parties, the desire to establish trust provides the intrinsic
motivation for individuals to consider other interests and curtail opportunistic behaviors. Trust development is often viewed
as an experiential learning process in which trustworthiness is
communicated to the other party through repeated social interactions [48], [53], [75], [82].
Client–consultant coordination offers a venue for such social
interactions. To the extent that the coordination effort is successful, it helps both organizations demonstrate to each other
their abilities to contribute to the project, thus establishing perceived trustworthiness. Ongoing coordination also creates a social group comprised of members from both organizations. Such
group memberships and associated positive experiences have
been found to promote trust [82]. Finally, face-to-face meetings
and other informal coordination help project team members become familiar with each other’s work processes which can foster
trust development as well [3]. Therefore, a trusting relationship
between client and consulting organizations can be fostered by
increased levels of formal and social interactions that are resulted from higher levels of coordination.
H3: Higher levels of interorganizational coordination will lead
to higher levels of trust between members of the client and
consultant organizations.
Project uncertainty is an important contingent factor that is
studied in project management [74]. Project uncertainty can be
260
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 57, NO. 2, MAY 2010
conceptualized as consisting of requirements uncertainty and
technical uncertainty [60]. Requirements uncertainty refers to
the vagueness about the application specifications desired by
the client organization, whereas technical uncertainty refers to
uncertainty associated with both the client and consulting organizations’ experience with the software and its implementation.
In internal IS projects, higher levels of coordination were found
to reduce the levels of project uncertainty [60]. In that study,
project uncertainty was measured by combining requirements
and technical uncertainty.
Since consultant-assisted IS projects have similar needs as
internal IS projects in determining project requirements, it is
expected that higher levels of coordination should have comparable impact on requirements uncertainty in consultant-assisted
projects. In addition, in consultant-assisted IS projects, consultants and clients bring different types of expertise to a
project team. Specifically, clients have the knowledge about
current business processes and needs. Sharing this knowledge
with the consultants, therefore, is critical to the project team
and its success [15]. Interorganizational coordination promotes
knowledge sharing by helping team members to recognize
the need for expertise, locate appropriate expertise, and apply expertise to address problems and issues [24]. Through
this coordination process, the team can effectively utilize the
business knowledge of clients to better determine application
requirements, thus reducing project requirements uncertainty.
Therefore,
H4: Higher levels of interorganizational coordination will lead
to lower levels of requirements uncertainty.
Technical uncertainty may lead to more complexity and ambiguity for the project work effort. Consultants often possess the
technical and project management skills that can reduce these
complexities and uncertainties. For the project to be successful,
the consultant’s knowledge must be shared with the project team.
Coordination can reduce technical uncertainty through the transfer of technical experience and knowledge from the consultants
to clients. As mentioned above, higher levels of coordination
were found to reduce the levels of project uncertainty, including
technical uncertainty, in internal IS projects [60]. We should
expect the same effects hold for consultant-assisted projects as
well. Therefore,
H5: Higher levels of interorganizational coordination will lead
to lower levels of technical uncertainty.
Both technical and requirements uncertainty have been found
to increase performance risks in software development [4],
[60], [61]. Das and Teng [16] suggest that perceived performance risk is inversely related to the trust of another’s ability
to carry out promised actions and achieve desired results. If
the client is able to clearly and consistently specify the application’s requirements with the help of consultants, the performance risks associated with the project should be reduced
significantly. Conversely, if the client is unclear about project
requirements and consultants are ineffective in helping clients
to better understand and specify them, misunderstanding and
confusion between client and consultant may arise. Such misunderstanding and confusion, in turn, can negatively impact
intention- and integrity-based trust [76]. Requirements uncertainty, often as a result of ambiguous and shifting project objectives, makes it more difficult for the client to communicate its
goals to the consultants and to design measures that ensure goal
congruence.
H6: Higher levels of project requirements uncertainty will lead
to lower levels of trust between the client and consultant organizations.
H7: Higher levels of project requirements uncertainty will lead
to lower levels of goal congruence between the client and consultant organizations.
Using a similar argument, when the client believes in the consultant’s ability to implement a particular technology, the perceived performance risk would likely be low. On the other hand,
if the technology is unfamiliar to the consultant, the client may
have concerns about project delivery. In addition, uncertainty
due to the lack of technical knowledge and frequent changes in
requirements may give rise to misunderstanding and confusion
between client and consultant concerning intention and project
objectives, and therefore negatively impact trust and goal congruence.
H8: Higher levels of project technical uncertainty will lead to
lower levels of trust between the client and consultant organizations.
H9: Higher levels of project technical uncertainty will lead to
lower levels of goal congruence between the client and consultant organizations.
Project uncertainty can also directly affect project performance, as [60] and [61] found for internal IS projects. Technical
uncertainty can increase the time and cost variances in completing a project if the team is unfamiliar with the technology.
Inexperience with the technology may also lead to implementation mistakes that can impact system quality. Similarly, project
requirements need to be clearly specified and agreed upon before
they can be implemented. Therefore, uncertainty and indecision
in requirement specifications could be a major source of project
delay, cost overrun, and dissatisfaction with the completed
system.
H10: Higher levels of project requirements uncertainty will lead
to lower levels of project performance.
H11: Higher levels of project technical uncertainty will lead to
lower levels of project performance.
From the economic exchange perspective, trust between organizations can reduce transaction costs [31], limit opportunistic behavior [8], and facilitate conflict resolution [65]. These
factors, in turn, have been shown to enhance the performance
of organizational transactions [84], such as those between the
clients and consultants. It follows that under such circumstances,
project performance is likely to be improved when high levels
of trust exist. Gefen [30] has shown that client’s trust of consulting organizations led to higher engagement success as measured by client satisfaction. Weber [81] found that client trust in
his or her project manager is positively related to project team
performance.
LIBERATORE AND LUO: COORDINATION IN CONSULTANT-ASSISTED IS PROJECTS: AN AGENCY THEORY PERSPECTIVE
H12: Higher levels of trust between members of the client and
consultant organizations will lead to higher levels of project
performance.
Goal congruence can be developed by identifying overlapping
and compatible goals among the partnering organizations [23].
To the extent that it can be achieved, goal congruence can impact
project performance in two significant ways. First, it can reduce
possible opportunistic behaviors induced by the agency problem
[37]. For example, to enhance their technical capabilities, the
consulting organization may have the tendency to experiment
with new technical features that may offer marginal benefits to
the client. Goal congruence, if it exists, will limit actions such
as these that are not consistent with the partners’ shared goals.
Second, both organizations have an incentive to put their efforts
in the same direction and thus can maximize the effects of joint
actions to achieve shared goals [37].
H13: Higher levels of goal congruence will lead to higher levels
of project performance.
IV. RESEARCH METHOD AND DATA ANALYSIS
We employed the survey methodology to test the proposed
model. In this section, we report on the instrument development,
data collection, and data analysis methods.
A. Instrument Development
A survey instrument was developed to test the hypotheses
stated above. The items used to measure the various constructs were developed from a review of the extant literature
to ensure content validity. Items measuring the level of coordination were developed based on the discussion of horizontal and vertical coordination and the items used by Nidumolu [60], [61]. Trust was measured using items that addressed
integrity, benevolence, and ability as adapted from Gefen [30],
who based his measures on the work of Mayer et al. [53].
Gefen’s [30] approach was followed, since he was addressing
the relationship between a client and a firm that customizes
ERP software. Items measuring the levels of requirements uncertainty, technical uncertainty, and goal congruence were developed based on the definitions of these concepts discussed
in the previous section. The level of project performance was
measured using items that reflected the quality of the project
management process (process quality), the quality of the application (product quality), and items that evaluated the extent
to which the project was over or under budgeted time and cost
[26].
The draft questionnaire was first reviewed by academics with
extensive experience in survey methodology, and was pretested
by a managing director of a major IT consulting firm. The survey
was then subjected to pilot testing using member organizations
on the CIO council of the authors’ business school. These organizations included manufacturing and service companies, as
well as major IT consulting firms. CIOs provided access to
project managers who participated in the pilot study. The questionnaire was modified and resulted in 33 items used to measure
six constructs, as shown in the Appendix.
261
B. Data Collection
The survey was conducted in cooperation with the Project
Management Institute (PMI), the largest project management
professional organization in the world (www.pmi.org). With the
assistance of the PMI research director, we identified the Information Systems and the Information Technology & Telecommunications special interest groups (SIGs) within PMI as the
target group for this study. We contacted the presidents of these
two SIGs and received their full support for our study. The SIGs
promoted our study and encouraged participation by including
a description and a prominent link to our survey on their secured
Web sites and in two electronic newsletters.
Following the link to the survey Web site, PMI SIG members
were first presented with a welcome page explaining the purpose
and value of the study. Key terms such as “project,” “application,” and “partner organization” were clearly defined. Next,
they were asked to focus on one IS implementation project in
which they had actively participated. In addition, the candidate
project should meet the following criteria: 1) have significant
participation of one or more external consulting organizations
that provide technical and/or project management expertise; 2)
involve the implementation of an enterprise system or an enterprise system module; and 3) be completed within the past 12
months or near completion. If the members did not have experience with such a project, they were directed to an exit page
where they were thanked for visiting the site. Otherwise, they
would enter the survey Web site and then complete the survey
questionnaire. As an incentive, all respondents were entered into
a lottery drawing for three Amazon gift certificates.
A total of 385 responses were received from the survey.
After removing duplicate submissions (hitting the submission
key multiple times) and responses with one or more sections
of missing data, we were left with 324 usable projects. Since
the percentage of PMI SIG members who had relevant project
experiences cannot be determined, computing a response rate
is problematic and it is possible that the sample may have a
bias, since respondents may be more likely to report successful projects [29]. However, in our sample, 61% of the reported
projects exceeded budgeted cost and 63% of them were not completed on time. These results are consistent with other surveys
reporting that a large portion of IS implementations are late and
over budget [43], [77]. In addition, with 324 projects, the sample
is quite large. As shown in Table I, the sample is also diverse
in application type, project size, and project length. Therefore,
we feel that this sample provides a good representation of the
population of consultant-assisted IS projects.
The project and respondent profiles for our sample are presented in Tables I and II, respectively. The projects reported in
the survey are composed of a variety of enterprise applications,
including complete ERP implementations and ERP modules
(financial, human resource, CRM, SCM, BI/KM, operations,
databases/infrastructure). Total budgeted costs of these projects
ranged from less than $100 000 to over $50 million, with the median cost falling into the $1–$5 million category. Project length
also varied greatly with a range of less than 3 months to over 24
months, with the median length falling into the 12–14 months
262
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 57, NO. 2, MAY 2010
TABLE I
PROJECT PROFILE
TABLE II
RESPONDENT PROFILE
category. About three-quarters of the projects were based in
the United States and Canada, with the remainder distributed
throughout the world. Table II shows that nearly one-half of
the respondents were project managers or program managers.
Almost one-third of the respondents were senior leaders, directors, and business managers, with the remainder composed of
business analyst/consultants and others, such as engineers and
information specialists. The average working experience of the
respondents was 17.5 years.
We examined possible sample bias due to project size and
location. Projects are divided into large projects (over 1 million
dollars in budget) and small projects. No significant differences
between large and small projects with respect to cost overrun
(p = 0.225) and time overrun (p = 0.318) were found. When
project location was divided into U.S.-based projects vs. nonU.S.-based projects, no significant differences were found for
cost (p = 0.229) and time (p = 0.126).
Since all the data are collected through the survey, common
method variance could be a potential source of bias. Harman’s
one-factor test is one of the widely used techniques in evaluating
the threat of common method biases in a data set [5], [18],
[64]. An exploratory factor analysis was performed on all items
measuring both independent and dependent variables. Seven
factors were extracted using the Kaiser criterion with the first
factor accounting for 28.48% of the total variance. The Harman’s
one-factor test result showed that there were multiple factors and
no one single factor could explain the majority of the variances
among variables. Therefore, common method variance is not a
cause for concern in our data set.
C. Data Analysis
Since some of the items have been adapted from previous
studies and others have not been tested before, exploratory
factor analysis was performed on the data set to assess the
initial convergent and discriminant validity. Items that loaded
on multiple factors or had a loading value of less than 0.40
were removed from further analysis. The collected dataset was
then analyzed and tested using the structural equation modeling approach to validate the proposed research model [7]. The
software package employed in this study was AMOS 4.0, a
covariance-based structural equation modeling tool. The testing of the proposed model involved the evaluation of both
the measurement model and the structural model. The purpose
of testing the measurement model is to determine if the measures possess satisfactory psychometric properties, i.e., do the
items in the questionnaire measure what they are supposed to
measure, while the structural model investigates the direction
and significance of causal relationships between various latent
variables. The psychometric properties of the “fit” measurement model were evaluated for convergent validity, discriminant validity, and overall model fit. When the measurement
LIBERATORE AND LUO: COORDINATION IN CONSULTANT-ASSISTED IS PROJECTS: AN AGENCY THEORY PERSPECTIVE
263
TABLE III
RELIABILITY MEASURES FOR CONSTRUCTS
TABLE IV
DISCRIMINANT VALIDITY
TABLE V
FIT INDEXES FOR THE MEASUREMENT MODEL AND STRUCTURAL MODEL
model is shown to have satisfactory psychometric properties, the
structural model can then be examined to validate the paths and
relationships.
V. RESULTS
Given the measurement model, psychometric properties of
various measures were assessed using confirmatory factor analysis (CFA). Table III reports the results of convergent validity
measures (construct reliabilities and average variance extracted
(AVE) by each construct) of the measurement model. All construct reliability values were above the threshold of 0.70. Most
AVE values were above the cutoff of 0.50, except for the coordination and goal congruence constructs. However, both constructs had rather high reliability measures (>0.80). Based on
the overall results, we found the convergent validity to be acceptable.
To show discriminant validity, the square root of AVE for
each construct should be greater than the correlation values of
that particular construct with all other constructs. As shown in
Table IV, this condition holds for all but one of the constructs.
Specifically, the correlation between coordination and goal congruence (0.654) was slightly larger than the square root of AVE
for coordination (0.650). Given this small difference, all of the
constructs within the measurement model are thought to have
acceptable discriminant validity. Substantial multicollinearity,
often indicated by high correlations (generally 0.80 or above)
among the independent variables, could also be a concern due
to its potential effect on estimation [32]. Table IV shows that the
correlations between coordination, trust, and goal congruence
ranged from 0.50 to 0.654, so multicollinearity does not pose a
significant threat to our model.
The measurement model fit indices are reported in Table V.
The normed χ2 was 1.728, which is below the recommended
cutoff value (less than 3.0). The adjusted goodness of fit index
was 0.867, which was above the threshold of 0.80. Both the
comparative fit index and Tucker–Lewis Index measures were
above the desired 0.90 recommended value. The root mean
square error of approximation was 0.047, which was below the
0.06 cutoff value. Only the Goodness of Fit Index of 0.893 was
slightly below the recommended cutoff value of 0.90. Overall,
the CFA results suggest that the proposed measurement model
fits the data and can be used to evaluate the relationships between
the constructs. The results of structural model fit indices are
reported in Table V and showed the same relationships as those
found for the measurement model.
The results of the analysis of the structural model are presented in Fig. 2, Tables VI and VII. Fig. 2 shows the standardized path coefficients and p-values between the constructs.
Requirements uncertainty was found to significantly and negatively impact trust (−0.15, p < 0.01) and goal congruence
(−0.315, p < 0.01), but its effect on project performance was
not significant (−0.145, p = 0.058). Similarly, technical uncertainty had significant negative effects on trust between the
client and consulting organizations (−0.219, p < 0.01) and goal
congruence (−0.16, p < 0.01), but its impact on project performance was not significant (−0.122, p = 0.08). Both trust
264
Fig. 2.
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 57, NO. 2, MAY 2010
Results of the structural model.
TABLE VI
RESULTS OF HYPOTHESIS TESTING
TABLE VII
VARIANCES EXPLAINED AND EFFECTS ON PROJECT PERFORMANCE
(0.309, p < 0.01) and goal congruence (0.283, p < 0.01) had
significant positive impacts on project performance. Coordination had significant and positive impacts on goal congruence
(0.588, p < 0.01) and trust (0.47, p < 0.01). On the other hand,
higher levels of coordination had significant effects in reducing requirements uncertainty (−0.238, p < 0.01) and technical
uncertainty (−0.173, p = 0.02). However, there was no significant direct effect of coordination on project performance
(0.152, p = 0.184). Table VI summarizes the acceptance or rejection of each hypothesis.
Table VII reports the breakdowns of direct and indirect effects of variables on project performance and the percentage of
LIBERATORE AND LUO: COORDINATION IN CONSULTANT-ASSISTED IS PROJECTS: AN AGENCY THEORY PERSPECTIVE
variance explained for each endogenous constructs. Trust and
goal congruence had the highest direct effects on project performance at 0.309 and 0.283, respectively. Coordination, however,
had the largest total effect on project performance at 0.572.
Its main impact on project performance was indirect, resulting
from its effects on goal congruence, trust, and requirements uncertainty. The combined effect of these five factors explained
53.9% of the variance in project performance. The variances
explained for goal congruence and trust were 52.1% and 36.2%,
respectively. The R2 values for technical uncertainty (4.4%)
and requirements uncertainty (9.2%) were comparatively low,
because they were based only on the correlations between coordination and uncertainty.
VI. DISCUSSION
The purpose of this study is to examine the impact of client–
consultant coordination level on the performance of consultantassisted IS projects. The results provide insights into our understanding of the interplay between coordination, uncertainty,
trust, and goal congruence, and their effects on project performance.
A. Contributions
This study shows that the level of client–consultant coordination has a large and significant total effect on project performance, but the direct effect is not significant. Coordination is
seen to influence performance primarily through the reduction
of technical and requirements uncertainty and the promotion
of trust and goal congruence. This result is interesting as it
differs from earlier studies of internal IS projects where coordination activities were found to directly affect project performance [60], [61]. Specifically, coordination was thought to
enable business users and IS staff to evaluate a fuller range of
options and produce a better quality system. A plausible reason for the difference is that the agency problem may be more
pronounced in projects involving multiple organizations. Consequently, in this context, significant coordination efforts are
directed to address uncertainty and agency-related issues. For
example, Sabherwal [68] found that the choice of coordination
mechanisms for outsourced software development projects was
influenced by the project uncertainty and the quality of the relationships between the parties. Similarly, the consultants and
clients in our study may have focused their coordination efforts on reducing requirements uncertainty, building trust, and
resolving goal conflicts as opposed to improving system quality and performance. The implication is that client–consultant
coordination is used primarily as a mechanism for managing
relationships.
Our results also show that better relationships can lead to
improved project performance. Specifically, trust and goal congruence were found to have significant positive impacts on
project performance. This finding suggests that a critical aspect of project management in consultant-assisted IS projects
is to effectively build trust and goal congruence between the
parties. Lacity and Hirschheim [45] proposed that the relationships between vendors and clients in IS outsourcing projects
265
should not be viewed as partnerships, because the overriding
profit motivation of the vendors would put the organizations in
direct conflict. However, our study shows that a partnership between the client and consulting organizations is possible if the
agency problem can be effectively addressed. Building trust and
goal congruence can help to resolve the agency problem in several ways. First, since a trusting relationship requires investment
from both parties, it can be the base upon which cooperation can
be developed. Second, trust can limit opportunistic behavior by
moderating consulting firms’ short-term profit motivation with
their long-term desire for future business and better reputation.
Third, by achieving goal congruence, both organizations have an
incentive to put their efforts in the same direction and maximize
overall project performance.
The research model provides guidance on how trust might
be developed. The notion that higher levels of client–consultant
coordination can lead to improved trust is supported by this research. The coordination process can lead to more transparent
work processes, enhanced social interactions, and knowledge
about each other’s capabilities. In addition, the results indicate
that reducing requirements and technical uncertainty could help
to build trust and goal congruence. Project uncertainty, including
both technical and requirements uncertainty, tends to increase
the project risks, leading to potential conflicts between the partners and reduced performance. Therefore, to the extent that
project uncertainty can be reduced, it will lessen project risks,
enhancing the opportunity for the partners to establish a trusting relationship and align respective goals. Furthermore, higher
levels of requirements and technical uncertainty are not found
to have significant direct impact on project performance. This
result differs from those reported in previous research of internal
IS projects [60], [61], indicating the uniqueness of consultantassisted projects and the importance of trust and goal congruence
in such projects.
This study makes several contributions to our understanding
of coordination and agency theory in the context of consultantassisted projects. First, our study highlights the significance
of the agency problem in consultant-assisted projects. While
research on improving internal IS project performance has focused on the effects of project characteristics, such as project
uncertainty, size, and risks, managing the agency relationship is
seen to be extremely important in consultant-assisted projects.
This finding supports efforts to further develop theories of trust
and goal congruence to address the agency problem. Second,
coordination has been identified as a key factor in mitigating
the agency problem in consultant-assisted projects, i.e., through
the coordination process, the parties can build relationships that
deepen trust and promote goal congruence.
Third, our research demonstrates an important linkage between coordination and agency theory. Previous studies on IS
projects have focused either on coordination theory [1], [60],
[61] or on agency theory [51] in the context of internal IS
projects. This study provides a fuller understanding of the role of
client–consultant coordination on project performance by clarifying its relationship with constructs developed from agency
theory, i.e., trust and goal congruence. Interorganizational coordination is shown to be an effective approach for addressing
266
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 57, NO. 2, MAY 2010
the agency problem by helping the parties to build relationships.
Overall, the contributions of this study represent a step toward
the development of a new theory on the role of interorganizational coordination.
The results of our study also provide valuable insights to practicing managers of both client and consulting organizations who
are engaged in consultant-assisted projects. The agency problem may have a more significant effect on project outcomes in
consultant-assisted projects than in internal IS projects. Therefore, client managers are advised to put more efforts, including
coordination, toward addressing possible agency-related problems. Our study shows that client managers should not only rely
on outcome-based contracts and monitoring but also focus on
building trusting relationships and seeking goal alignment with
their consulting partners.
From the consultant perspective, a trust-based relationship
can also be beneficial for two important reasons. First, being
a trusted partner can give consultants a competitive advantage
for obtaining future contracts with the same client, since related
negotiation, transaction, and control costs could be significantly
reduced for both parties [30]. Second, consulting organizations
rely on their reputation for long-term success. Trust-based partnerships can lead to higher client satisfaction and in turn enhance
consultant’s reputation in the marketplace.
ory, and their interrelationships impact project performance, it
can be augmented by including specific coordination mechanisms, additional agency theory constructs, and interpersonal
trust measures. First, the effects of specific coordination mechanisms and activities were not considered in the model. As a
result, we cannot make recommendations as to what specific
coordination mechanisms should be employed and under what
circumstance to improve performance. However, regardless of
the specific coordination mechanisms used, our study highlights
the importance of coordination in consultant-assisted projects.
Second, the agency constructs considered in this research were
limited to trust and goal congruence. By incorporating other important agency theory constructs, such as formal contracts and
monitoring, we would be able to compare the effects of various
mechanisms for addressing agency problems and examine the
tradeoffs between them.
Third, the measurement of trust was limited to that occurring
at the interorganizational level between the partner organizations. As a result, we cannot examine the effects of interpersonal
trust on project performance and its relationship with interorganizational trust. Understanding the differences and relationships
between interpersonal and interorganizational trust can help us
explain how the latter can be evolved and developed [84]. Each
of the three areas presents a promising direction for future research.
B. Limitations and Future Research
Even though our sample size was large, the survey was conducted using a single informant for each project. This is an important limitation of our research. An alternate approach would
involve obtaining multiple responses from each project team.
The latter approach would not only increase survey reliability but would also allow us to determine if there are different
perspectives among team participants within the same project.
However, the tradeoff is the difficulty of obtaining a sufficiently
large-scale and cross-sectional sample for model validation. In
addition to the survey instrument, a follow-up, semistructured
interview of the respondent would no doubt yield further insights about the projects and clarify issues such as whether the
levels of coordination, trust, and goal congruence evolve during
the course of the project. The insight gained through that process
would serve as an informative and important complement to our
study. Such an effort can form the basis of future research.
Client and consulting organizations may have varying degrees of interest in a range of project performance measures.
We have focused on project performance measures, which are
primarily of interest to the client. Profitability, client satisfaction, knowledge acquisition, and potential for future business are
examples of performance measures that are of special interest to
consultants. It is conceivable that the relationships in our model
might be altered if these measures are considered. Therefore,
by including performance measures of interest to all parties, we
would be able to generalize the model to fully consider their
joint interest.
There are several additional opportunities for future research
that follow from this study. While the model provides a highlevel understanding of how coordination theory, agency the-
VII. SUMMARY
This study developed a research model for examining the
effects of the level of client–consultant coordination on the performance of consultant-assisted IS projects by integrating the
coordination and agency theory perspectives. We hypothesized
that trust, goal congruence, and project uncertainty were the mediating factors between coordination and project performance.
While the proposed model is generally supported by the data
collected for the study, coordination was not found to have a
direct effect on project performance. Instead, coordination has
an indirect effect by reducing uncertainty and building trust
and goal congruence. This research contributes to the theory
and practice in IS project management by clarifying the role
of coordination, demonstrating the importance of the agency
relationship in consultant-assisted projects, suggesting ways to
effectively manage this relationship, and providing a framework
for further studying consultant-assisted IS projects.
APPENDIX
Coordination
C1) Designated individual(s) within the project team (e.g., a
project manager) was/were responsible for coordination
with the partner organization
C2*) A designated committee or group (e.g., a steering committee) outside of the project team was responsible for
coordination with the partner organization
C3) Our project team worked interactively with our partners
on important aspects of the project
LIBERATORE AND LUO: COORDINATION IN CONSULTANT-ASSISTED IS PROJECTS: AN AGENCY THEORY PERSPECTIVE
C4*) Coordination between partner organizations was performed on an as-needed basis
C5) Giving instructions or directions is an important way we
coordinate with our partners
C6) There were established rules and procedures for coordination between the partner organizations
C7) Team members were engaged in coordinating with our
partners
C8*) I coordinated more with members of our project team
than with my manager
C9) I worked interactively with my counterpart from the partner organization on important aspects of the project
C10) Coordination between partner organizations took place at
the project manager/leader level
267
Project Performance
PP1) The quality of the project management is excellent
PP2) We are completely satisfied with the application implemented in the project
PP3) By approximately what percentage, if any, did actual
costs for the project overrun originally budgeted costs?
(indicate underrun by negative sign)
PP4) By approximately what percentage, if any, did actual
completion time for the project overrun originally budgeted completion time? (indicate underrun by negative
sign)
∗
items removed after the exploratory factor analysis
REFERENCES
Requirements Uncertainty
RU1*) The application requirements of the project were quite
different for different groups of users
RU2) The application requirements were well known at the
outset of the project (Reversed)
RU3) The application requirements changed quite a bit during
the project
RU4) The application requirements identified at the beginning
of the project were quite different from those existing at
the end
Technical Uncertainty
TU1) The project team was inexperienced with the software
package
TU2) The project team had implemented several similar systems before
TU3) The project team was familiar with the software package
implementation (Reversed)
Trust
T1*) Our partners are competent in their field
T2*) Our partners are knowledgeable concerning the application
T3*) Our partners put their interests before ours
T4) Our partners are honest about their problems
T5) Our organization can count on our partners to be sincere
T6) We can count on our partners to consider how their
decisions will affect us
T7) Our partners are open in dealing with us
Goal Congruence
GC1) The goals of our partners regarding this project were
clearly communicated to our project team
GC2) Our partners were asked to achieve goals that were not
the same as ours (Reversed)
GC3) We clearly communicated our goals to our partners
GC4) The goals of our partners were consistent with our goals
GC5) Our partners agree with us on the priority of the goals
to be achieved in this project
[1] H. P. Andres and R. W. Zmud, “A contingency approach to software
project coordination,” J. Manage. Info. Syst., vol. 18, no. 3, pp. 41–72,
2001.
[2] R. D. Austin, “The Effects of time pressure on quality in software development: An agency model,” Inform. Syst. Res., vol. 12, no. 2, pp. 195–207,
2001.
[3] R. Bachmann, “The coordination of relations across organizational boundaries,” Int. Stud. Manag. Org., vol. 33, no. 2, pp. 7–21, 2003.
[4] H. Barki, S. Rivard, and J. Talbot, “Toward an assessment of software
development risk,” J. Manag. Inform. Syst., vol. 10, no. 2, pp. 203–225,
1993.
[5] A. Barua, P. Konana, A. B. Whinston, and F. Yin, “An empirical investigation of net-enabled business value,” MIS Quart., vol. 28, no. 4,
pp. 585–617, 2004.
[6] C. Berggren, J. Soderlund, and C. Anderson, “Clients, contractors, and
consultants: The consequences of organizational fragmentation in contemporary project environments,” Project Manag. J., vol. 32, no. 3, pp. 39–48,
2001.
[7] K. A. Bollen, Structural Equations With Latent Variables. New York:
Wiley, 1989.
[8] P. Bromiley and L. L. Cummings, “Transaction costs in organizations with
trust,” in Research on Negotiation in Organizations, R. Bies, B. Sheppard,
and R. Lewicki, Eds. Greenwich, CT: JAI Press, 1995.
[9] C. Brown and I. Vessey, “Managing the next wave of enterprise systems:
Leveraging lessons from ERP,” MIS Quart. Exec., vol. 2, no. 1, pp. 65–77,
2003.
[10] E. Carmel and S. Sawyer, “Packaged software development teams: What
makes them different?,” Inform. Technol. People, vol. 11, no. 1, pp. 7–19,
1998.
[11] R. Casadesus-Masanell, “Trust in agency,” J. Econ. Manag. Strategy,
vol. 13, no. 3, pp. 375–404, 2004.
[12] F. Chiocchio, “Project team performance: A study of electronic task and
coordination communication,” Project Manag. J., vol. 38, no. 1, pp. 97–
109, 2007.
[13] T. Clark, R. W. Zmud, and G. McCray, “The outsourcing of information
services: Transforming the nature of business in the information industry,”
in Strategic Sourcing of Information Systems, L. P. Willcocks and M.
C. Lacity, Eds. New York: Wiley, 1998.
[14] J. G. Cooprider and J. C. Henderson, “Technology-process fit: Perspectives
on achieving prototyping effectiveness,” J. Manag. Inform. Syst., vol. 7,
no. 3, pp. 67–87, 1991.
[15] J. Cummings, “Work groups, structural diversity, and knowledge sharing
in a global organization,” Manag. Sci., vol. 50, no. 3, pp. 352–364, 2004.
[16] T. K. Das and B. Teng, “The risk-based view of trust: A conceptual
framework,” J. Bus. Psychol., vol. 19, no. 1, pp. 85–116, 2004.
[17] W. H. DeLone and E. R. McLean, “Information systems success: The quest
for the dependent variable,” Inform. Syst. Res., vol. 3, no. 1, pp. 60–95,
1992.
[18] S. Devaraj, M. Fan, and R. Kohli, “Antecedents of B2C channel satisfaction and preference: Validating e-commerce metrics,” Inform. Syst. Res.,
vol. 13, no. 3, pp. 316–333, 2002.
[19] M. Doane, The SAP Bluebook: A Concise Business Guide to the World of
SAP. Evans Mills, NY: Doane Associates Press, 1998.
[20] W. J. Doll and G. Torkzadeh, “The measurement of end user computing
satisfaction,” MIS Quart., vol. 12, pp. 259–274, 1988.
268
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. 57, NO. 2, MAY 2010
[21] L. Donaldson, “The ethereal hand: Organizational economics and management theory,” Acad. Manag. Rev., vol. 15, pp. 369–381, 1990.
[22] K. Eisenhardt, “Agency theory: An assessment and review,” Academy
Manag. Rev., vol. 14, no. 1, pp. 57–74, 1989.
[23] J. Eliashberg and D. A. Michie, “Multiple business goal sets as determinants of marketing channel conflict: An empirical study,” J. Market. Res.,
vol. 21, pp. 75–84, 1984.
[24] S. Faraj and L. Sproull, “Coordinating expertise in software development
teams,” Manag. Sci., vol. 46, no. 12, pp. 1554–1568, 2000.
[25] L. M. Farrell, “Principal-agency risk in project finance,” Int. J. Project
Manag., vol. 21, pp. 547–561, 2003.
[26] G. G. Gable, “A multidimensional model of client success when engaging
consultants,” Manag. Sci., vol. 42, no. 8, pp. 1175–1198, 1996.
[27] J. R. Galbraith, Designing Complex Organizations.
Reading, MA:
Addison-Wesley, 1973.
[28] Gartner, Group, Forecast: IT Services, Worldwide, 2005—2009. Stamford, CT: Gartner, Inc., 2005.
[29] T. F. Gattiker and D. L. Goodhue, “What happens after ERP implementation: Understanding the impact of interdependence and differentiation on
plant-level outcomes,” MIS Quart., vol. 29, no. 3, pp. 559–585, 2005.
[30] D. Gefen, “Nurturing clients’ trust to encourage engagement success during the customization of ERP systems,” Omega, vol. 30, pp. 287–299,
2002.
[31] R. Gulati, “Does familiarity breed trust? The implications of repeated ties
for contractual choice in alliances,” Academy Manag. J., vol. 38, no. 1,
pp. 85–112, 1995.
[32] F. H. Hair, R. E. Anderson, R. L. Tatham, and W. C. Black, Multivariate
Data Analysis, 5th ed. Upper Saddle River, NJ: Prentice-Hall, 1998.
[33] F. Hartman, Ten Commandments of Better Contracting: A Practice Guide
to Adding Value to an Enterprise Through More Effective SMART Contracting. Reston, VA: ASCE Press, 2003, pp. 235–260.
[34] K. Hong and Y. Kim, “The critical success factors for ERP implementation:
An organizational fit perspective,” Inform. Manag., vol. 40, no. 1, pp. 25–
40, 2002.
[35] L. T. Hosmer, “Trust: The connecting link between organizational theory
and philosophical ethics,” Academy Manag. Rev., vol. 20, no. 2, pp. 379–
403, 1995.
[36] S. D. Jap, “Pie-expansion efforts: Collaboration processes in buyersupplier relationships,” J. Marketing Res., vol. 36, no. 4, pp. 461–475,
1999.
[37] S. D. Jap and E. Anderson, “Safeguarding international performance and
continuity under ex post opportunism,” Manag. Sci., vol. 49, no. 12,
pp. 1684–1701, 2003.
[38] K. N. Jha and K. C. Iyer, “Critical determinants of project coordination,”
Int. J. Project Manag., vol. 24, pp. 314–322, 2006.
[39] K. N. Jha and K. C. Iyer, “Commitment, coordination, competence and
the iron triangle,” Int. J. Project Manag., vol. 25, pp. 527–540, 2007.
[40] A. Kadefors, “Trust in project relationships—inside the black box,” Int.
J. Project Manag., vol. 22, pp. 175–182, 2004.
[41] S. L. Kirby and M. A. Davis, “A study of escalating commitment in
principal-agent relationships: Effects of monitoring and personal responsibility,” J. Appl. Psychol., vol. 83, no. 2, pp. 206–217, 1998.
[42] L. J. Kirsch, V. Sambamurthy, D. Ko, and R. L. Purvis, “Controlling
information systems development projects: The view from the client,”
Manag. Sci., vol. 48, no. 4, pp. 484–498, 2002.
[43] C. Koch (2002, Jul. 15). It’s Time to Take Control. CIO Mag.
[44] R. E. Kraut and L. A. Streeter, “Coordination in software development,”
Commun. ACM, vol. 38, no. 3, pp. 69–81, 1995.
[45] M. C. Lacity and R. Hirschheim, Information Systems Outsourcing.
Chichester, U.K.: Wiley, 1993.
[46] W. M. Lassar and J. L. Kerr, “Strategy and control in supplier-distributor
relationships: An agency perspective,” Strategic Manag. J., vol. 17, no. 8,
pp. 613–632, 1996.
[47] R. J. Lewicki and B. B. Bunker, “Trust in relationships: A model of
development and decline,” in Conflict, Cooperation and Justice: Essays
Inspired by the Work of Morton Deutsch, B. B. Bunker and J. Z. Rubin,
and Associates, Eds. San Francisco, CA: Jossey-Bass, 1995, pp. 133–
173.
[48] R. J. Lewicki and B. B. Bunker, “Developing and maintaining trust in
working relationships,” in Trust in Organizations: Frontiers of Theory
and Research, R. M. Kramer and T. R. Tyler, Eds. London, U.K.: Sage,
1996, pp. 114–139.
[49] R. J. Lewicki, E. C. Tomlinson, and N. Gillespie, “Models of interpersonal
trust development: Theoretical approaches, empirical evidence, and future
directions,” J. Manag., vol. 32, pp. 991–1022, 2006.
[50] M. Liberatore and W. Luo, “The effect of client—consultant coordination
on is project performance: An agency theory perspective,” presented at
the PICMET 2007, Portland, OR, Aug. 5–9.
[51] R. C. Mahaney and A. L. Lederer, “Information systems project management: An agency theory interpretation,” J. Syst. Softw., vol. 68, pp. 1–9,
2003.
[52] T. W. Malone and K. Crowston, “The interdisciplinary study of coordination,” ACM Comput. Surv, vol. 26, no. 1, p. 87, 1994.
[53] R. C. Mayer, J. H. Davis, and F. D. Schoorman, “An integrative model of
organizational trust,” Academy Management Rev., vol. 20, no. 3, pp. 709–
734, 1995.
[54] D. J. McAllister, “Affect- and cognitive-based trust as foundations for
interpersonal cooperation in organizations,” Academy of Management
Journal, vol. 38, no. 1, pp. 24–59, 1995.
[55] F. W. McFarlan and R. L. Nolan, “How to manage an outsourcing alliance,”
Sloan Manag. Rev., vol. 36, no. 2, pp. 9–23, 1995.
[56] D. H. McKnight, L. L. Cummings, and N. L. Chervany, “Initial trust
formation in new organizational relationships,” Academy Manag. Rev.,
vol. 23, no. 2, pp. 472–490, 1998.
[57] R. Mirani, “Procedural coordination and offshored software tasks: Lessons
from two case studies,” Inform. Manag., vol. 44, pp. 216–230, 2007.
[58] T. Moynihan, “Coping with client-based “people-problems”: The theoriesof-action of experienced IS/software project managers,” Inform. Manag.,
vol. 39, no. 2, pp. 377–390, 2002.
[59] R. Muller and J. R. Turner, “The impact of principal–agent relationship
and contract type on communication between project owner and manager,”
Int. J. Project Manag., vol. 23, pp. 398–403, 2005.
[60] S. R. Nidumolu, “The effect of coordination and uncertainty on software
project performance: Residual performance risk as an intervening variable,” Inform. Syst. Res., vol. 6, no. 3, pp. 191–219, 1995.
[61] S. R. Nidumolu, “A comparison of the structural contingency and riskbased perspectives on coordination in software-development projects,” J.
Manag. Inform. Syst., vol. 13, no. 2, pp. 77–113, 1996.
[62] N. Parolia, S. Goodman, Y. Li, and J. J. Jiang, “Mediators between coordination and is project performance,” Inform. Manag., vol. 44, pp. 635–645,
2007.
[63] J. K. Pinto and D. P. Slevin, “Critical success factors in effective project
implementation,” in Project Management Handbook, D. I. Cleland and
W. R. King, Eds. New York: Van Nostrand Reinhold, 1988, pp. 479–
512.
[64] P. M. Podsakoff, S. B. MacKenzie, J.-Y. Lee, and N. P. Podsakoff, “Common method biases in behavioral research: A critical review of the literature and recommended remedies,” J. Appl. Psychol., vol. 88, no. 5,
pp. 879–903, 2003.
[65] P. S. Ring and A. H. Van de Ven, “Developmental processes of cooperative
interorganizational relationships,” Academy Manag. Rev., vol. 19, no. 1,
pp. 90–118, 1994.
[66] S. Rosen, “Contracts and the market for executives,” in Contract Economics, L. Werin and H. Wijkander, Eds. Cambridge, MA: Blackwell,
1992, pp. 181–211.
[67] D. M. Rousseau, S. B. Sitkin, R. S. Burt, and C. Camerer, “Not so different
after all: A cross-discipline view of trust,” Academy Manag. Rev., vol. 23,
no. 2, pp. 393–404, 1998.
[68] R. Sabherwal, “The evolution of coordination in outsourced software development projects: A comparison of client and vendor perspectives,”
Inform. Org., vol. 13, pp. 153–202, 2003.
[69] S. Sawyer, “Packaged software: Implications of the differences from custom approaches to software development,” Eur. J. Inform. Syst., vol. 9,
no. 1, pp. 47–58, 2000.
[70] S. Sawyer, “A market-based perspective on information systems development,” Commun. ACM, vol. 44, no. 11, pp. 97–102, 2001.
[71] S. M. Schmidt and T. A. Kochan, “Interorganizational relationships: Patterns and motivations,” Admin. Sci. Quart., vol. 22, no. 2, pp. 220–234,
1977.
[72] J. E. Scott and I. Vessey, “Managing risks in enterprise systems implementations,” Commun. ACM, vol. 45, no. 4, pp. 74–81, 2002.
[73] S. P. Shapiro, “Agency Theory,” Annu. Rev. Sociol., vol. 31, no. 4, pp. 263–
284, 2005.
[74] A. J. Shenhar, “One size does not fit all projects: Exploring classical
contingency domains,” Manag. Sci., vol. 47, no. 3, pp. 394–414, 2001.
[75] B. H. Sheppard and D. M. Sherman, “The Grammars of trust: A model and
general implications,” Academy Manag. Rev., vol. 23, no. 3, pp. 422–437,
1998.
[76] R. M. Sorrentino, J. G. Holmes, S. E. Hanna, and A. Sharp, “Uncertainty orientation and trust in close relationships: Individual differences
LIBERATORE AND LUO: COORDINATION IN CONSULTANT-ASSISTED IS PROJECTS: AN AGENCY THEORY PERSPECTIVE
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
in cognitive styles,” J. Pers. Soc. Psychol., vol. 68, no. 2, pp. 314–327,
1995.
Standish Group International. (2004). CHAOS Report. Boston, MA, 2010.
H. L. Tosi, J. P. Katz, and L. R. Gomez-Mejia, “Disaggregating the agency
contract: The effects of Monitoring, incentive alignment, and term in office
on agent decision making,” Academy Manag. J., vol. 40, no. 4, pp. 584–
602, 1997.
J. R. Turner and R. Muller, “On the nature of the project as a temporary
organization,” Int. J. Project Manag., vol. 21, pp. 1–8, 2003.
A. H. Van De Ven, A. L. Delbecq, and R. Koenig, Jr., “Determinants of
coordination modes within organizations,” Amer. Sociol. Rev., vol. 41,
no. 2, pp. 322–338, 1976.
S. S. Weber, “Blending service provider-client project teams to achieve
client trust: Implications for project team trust, cohesion, and performance,” Project Manag. J., vol. 39, no. 2, pp. 72–81, 2008.
M. Williams, “In whom we trust: Group membership as an effective
context for trust development,” Academy Manag. Rev., vol. 26, no. 3,
pp. 377–396, 2001.
P. S.-P. Wong and S.-O. Cheung, “Trust in construction partnering: Views
from parties of the partnering dance,” Int. J. Project Manag., vol. 22,
pp. 437–446, 2004.
A. Zaheer, B. McEvily, and V. Perrone, “Does trust matter? Exploring
the effects of interorganizational and interpersonal trust on performance,”
Org. Sci., vol. 9, no. 2, pp. 141–159, 1998.
R. Zaghloul and F. Hartman, “Construction contracts: The cost of mistrust,” Int. J. Project Manag., vol. 21, no. 6, pp. 419–424, 2003.
Matthew J. Liberatore received the B.A. degree in
mathematics, and the M.S. and Ph.D. degrees in operations research from the University of Pennsylvania,
Philadelphia.
He is currently the John F. Connelly Chair in Management and the Director of the Analytics Strategic
Initiative Group, Villanova School of Business, Villanova University, Villanova, PA. He has authored extensively in the fields of management science, project
management, information systems, and research and
engineering management. His current research interests include project management planning and scheduling, information and
technology project management, and decision support systems for health care
administration.
Dr. Liberatore is a member of the Editorial Boards of the IEEE Transactions
on Engineering Management, American Journal of Mathematical and Management Sciences, Computers and Operations Research. He is a member of the
Decision Sciences Institute, the Institute for Operations Research and the Management Sciences, and the Project Management Institute.
View publication stats
269
Wenhong Luo received the B.Sc. degree in mathematics from East China Normal University, Shanghai, China, the M.B.A. degree from Nyenrode Business Universiteit, Breukelen, The Netherlands, and
the Ph.D. degree in business administration from the
University of Kentucky, Lexington.
He is currently an Associate Professor in the Department of Accounting and Information Systems,
Villanova School of Business, Villanova University,
Villanova, PA. His current research interests include
information systems project management, business
process management, and business intelligence. His publications have appeared
in leading journals such as Communications of the ACM, European Journal of
Information Systems, Information and Management, IEEE Transactions on
Engineering Management, International Journal of Production Research,
Omega, and the Information Society.
Keep reading this paper — and 50 million others — with a free Academia account
Used by leading Academics
Kenneth Pakenham
The University of Queensland, Australia
Irina Malkina-Pykh
Saint-Petersburg State University
Thomas L Webb
The University of Sheffield
Eric S . Kim (UBC)
University of British Columbia
