Content uploaded by Suling Zhang
Author content
All content in this area was uploaded by Suling Zhang
Content may be subject to copyright.
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 47
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
ABSTRACT
Virtual teams are an important work structure in software development projects. However, little is known
about what constitutes effective virtual software team leadership, in particular, the amount of leader del-
egation that is appropriate in a virtual software-development environment. This study investigates virtual
software team leader delegation and explores the impact of delegation strategies on virtual team performance
mediated by team motivation, team exibility and team satisfaction with the team leader. This research is a
report of a pilot study run on student teams carried out to rene and test the research constructs and research
model for a larger study run in corporations. The study found that virtual team leaders delegate more to
competent virtual teams and that such delegation is positively correlated with team member satisfaction
with their leader and with team member motivation. Overall, the work provides important information for
software-based organizations interested in developing virtual team leadership skills.
Keywords: delegation strategies, leadership, virtual software teams
INTRODUCTION
Virtual teams are composed of geographically
distributed coworkers linked though informa-
tion technologies to achieve an organizational
task (Townsend, DeMarie, & Hendrickson,
1998). Virtual teams are a popular structure in
software development for several reasons: they
provide access to lower-cost labor as well as
access to a range of disciplines and technical
specialties (Curtis, Krasner, & Iscoe, 1988).
While software team leaders and managers are
Occurrence and Effects of
Leader Delegation in Virtual
Software Teams
Suling Zhang, Kean University, USA
Marilyn Tremaine, New Jersey Institute of Technology, USA
Rich Egan, New Jersey Institute of Technology, USA
Allen Milewski, Monmouth University, USA
Patrick O’Sullivan, IBM Dublin Lab, Ireland
Jerry Fjermestad, New Jersey Institute of Technology, USA
IGI PUBLISHING
This paper appears in the publication, International Journal of e-Collaboration , Volume 5, Issue 1
edited by Ned Kock © 2008, IGI Global
701 E. Chocolate Avenue, Suite 200, Hershey PA 17033-1240, USA
Tel: 717/533-8845; Fax 717/533-8661; URL-http://www.igi-global.com
ITJ 4626
48 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
now frequently given virtual teams to manage,
they have not been given clear directions on
how to effectively manage such teams. One
important issue regarding virtual software team
management is when and how team leaders
should delegate authority and responsibility
to the team.
Delegation means that one has been em-
powered by one’s superior to take responsibility
for certain activities, which were originally
reserved for the superior (Ashton & Kramer,
1980; Bass, 1990). In traditional leadership
studies, delegation is widely acknowledged to
be an essential element of effective manage-
ment (Yukl, 2002), and effective delegation
offers a number of potential benets, both to
the manager and the subordinates. However,
to the authors’ knowledge, only a few studies
have been conducted to investigate delegation
as a distinct component of global virtual team
leadership. In the limited number of conceptual
works and empirical studies in which delegation
is not the direct focus, delegation has been a
controversial issue. Some researchers argue for
the benets of delegation: Eveland and Bikson,
(1988); Jarvenpaa, Knoll, and Leidner, (1998);
Jarvenpaa and Leidner, (1999) report that an
effective leader of a virtual team needs to be
more exible to accommodate the complexities
and volatility of the virtual team environment,
and to be willing to let others take the lead
when necessary. Furthermore, they suggest
that virtual team leadership should focus on
facilitating and empowering team members
to take action on their own. In contrast, Paré
and Dubé, (1999) argue that, due to the dis-
tributed nature of virtual teams, management
by observation is simply not possible, and that
much more discipline and control is required in
a virtual setting. Additionally, team effective-
ness in virtual environments may be hindered
by excessive autonomy coupled with exclusive
reliance on electronic communication and lack
of face-to-face interaction.
This article aims to address the research gap
regarding delegation to a virtual software team
by investigating the occurrence and effects of
leader delegation in such teams. Finding out how
and why leaders do or do not delegate to virtual
teams and the impact of the leaders’ delega-
tion behaviors on the teams will help industry
practitioners to better frame their strategy for
managing distributed software teams and will
also add to the eld’s knowledge about virtual
team leadership.
The focus of the article is on software teams,
in part, because it is felt that the global software
team phenomenon has several unique charac-
teristics that may not apply to virtual teams in
general. Unlike other activities that have been
outsourced or offshored, work activities cannot
be as easily compartmentalized because of the
high integration of the software product. Thus,
there is a need for communication and working
together in a structured fashion, which demands
good leadership. In addition, software develop-
ers expect to have a high degree of independence
in their work. Therefore, the degree of delega-
tion by a leader may differentially affect these
virtual teams.
Collaboration in software development also
demands the ability to communicate highly de-
tailed specications and questions. This requires
communication skills that may not be needed in
other types of virtual teams, skills which may
be inuenced by a team member’s knowledge
of English, the common language for many of
these teams. Therefore, good management of
the communication structure and media is likely
to be an important leadership trait. Finally, the
new countries that are now being included in
global software development have relatively
young team members. As such, the distribution
of corporate knowledge and software skills is
uneven. Thus, leadership and delegation in this
type of environment is likely to be different
for software than for other tasks carried out
by virtual teams.
The article is structured as follows: rst, an
overview of the research model is presented to
give readers a sense of the focus of the article;
then, based on a review of the literature, concep-
tualizations of leader delegation are presented
and specic research hypotheses regarding
virtual team leader delegation are explained;
nally, the pilot study testing the model is
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 49
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
presented. A discussion section presents the
contributions and limitations of this research,
and a nal section discusses the implications
for virtual team management.
RESEARCH MODEL
Before beginning a detailed discussion of the
variables used in this study, we present the
overall research model and briey describe
the relationships that are hypothesized to exist
between the variables. A key focus of the model
is the amount of delegation that virtual software
teams receive from their leaders. It behooves us,
therefore, to describe the management structures
that are being studied.
When we talk about virtual teams, the
structure that is typically in place does not
constitute virtuality for each and every team
member but, rather, distributed teams in which
some subsection of the software development
team is co-located and other sections are virtual.
For example, a portion of the team may be
located in the United States and a second and
third portion in India and China, respectively.
The overall team is working on the same soft-
ware product, but the work has typically been
compartmentalized in some way so that each
co-located portion of the team has specic
assignments. However, the work is such that
there is a need for continued communication
between each of the non-co-located portions of
the team to resolve integration issues.
Management of these teams comes from
several sources. First, there is typically a local
manager who handles personnel issues. Then,
there is a technical manager who oversees the
project. Finally, there is a technical head of
the particular subsection of the project who
is directing the work of both virtual and co-
located sub teams. Management of the work
is the purview of this leader, and this is where
delegation typically occurs. This management
is located with one sub team and virtual to the
other sub teams. Traditionally, management is
located in the home of the parent company, but
it also may be at a customer site.
Because at least, one part of the software
team is distant from the manager, less is likely
to be known about the competence of that part
of the team by the manager, and therefore, he
or she is likely to delegate in a manner so that
majority control of the technical issues rests
with that portion of the team that is co-located
with the manager. The research model being put
forward in this paper suggests that perceived
team competence predicts the amount of del-
egation that will occur but that this delegation
will have an effect on key team variables such
as satisfaction, motivation and exibility. It is
argued that these three variables are primary in
affecting team performance. Figure 1 presents
our research model.
An elemental part of this model is that team
competence not only affects the decision of the
team leader on whether to delegate or not, but
also affects the impact of the delegation. That
is, teams with less competence, are likely to
desire less delegation and to be unhappy and
unmotivated with less direction from the leader.
This occurs primarily because the team mem-
bers do not know how to perform their tasks
completely, yet wish to succeed in these tasks.
However, the only way they can succeed is if
they receive tighter direction from the leader.
As we shall see with our analysis, we did not
necessarily nd that less competent teams
wanted more delegation. The measure of per-
ceived competence that we used had difculties
with the student teams used in this research.
All student teams perceived themselves to be
highly competent. Rather than treating this as
a problem, it is a result that we believe has an
impact on virtual team management.
Virtual subteams are not able to ascertain
their competence in comparison to their distant
counterparts because they lack the proximity
required for an accurate comparison. They
are, therefore, likely to perceive themselves as
competent and desire more delegation. Culture
may also have an impact in that many cultures
worry about “face,” that is, how they appear
to others. These cultures would also perceive
themselves as competent, possibly due to team
members’ advanced education in their country
50 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
or membership in a high social class. What-
ever the reason, they are likely to desire more
delegation even when it is unwarranted. This
issue will be further addressed in the paper’s
discussion section.
The next sections present prior research on
the relations posited in our model and detailed
denitions of each of the variables used in the
model plus support from the literature on the
validity of the constructs we have developed
for the variables.
LITERATURE AND
HYPOTHESES
Delegation
As mentioned earlier, delegation is a state or
condition in which one has been empowered by
one’s superior to take responsibility for certain
activities, which were originally reserved for
the superior (Ashton & Cramer, 1980). There is
a rich body of studies in traditional leadership
research investigating delegation, mostly as a
feature of leadership style or as a combination
of related leader behaviors. Very few studies in-
vestigated delegation as a distinct management
practice (Yukl & Fu, 1999; Zhang, Tremaine,
Fjermestad et al., 2006). Because of this, we
cannot ascertain the effect size of delegation (as
a distinct leadership component) on possible
variables of interest.
Another major limitation of previous
delegation and leadership studies is the failure
to focus on what types of responsibilities and
activities a leader delegates. Only the global
delegative style of the leader was assessed. This
limitation seriously undermines the practical
value of these studies, as their ndings do not
inform software team managers where they can
and cannot delegate nor does it tell them what
they can delegate. Leadership is a multi-faceted
process, particularly so in virtual teams, given
the technological, cultural and organizational
complexities of a virtual team environment.
Therefore, it is important to examine the lead-
ership and managerial functions originally as-
signed to leaders to determine which of these can
be delegated to followers. To that end, studies
that describe leadership and managerial func-
tions are reviewed in this section, and a list of
leader delegation categories is generated.
For classical management theorists like
Davis (1942) and Urwick (1952), the functions
of the manager-leader in a formal organiza-
tion were orderly planning, organizing and
controlling. To address the overlapping needs
of the organization, team and individuals,
Figure 1. Proposed model of delegation effects on global software teams
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 51
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
Cofn (1944) modied the classical functions
as follows: formulation (planning), execution
(organizing), and supervision (persuading).
MacKenzie (1969) proposed a well-known
leader-manager model that illustrated the great
variety of activities that a typical manager per-
forms. He proposed that the central management
functions relate to the management of people,
ideas and things, which form the three basic
components of every organization with which
managers must work. Three functions (the
analysis of problems, decision-making and com-
munications) are important at all times and in all
aspects of the jobs held by managers, and there-
fore permeate the entire work process. To carry
out these functions, the leader-manager needs
to execute these leader activities: planning,
organizing, stafng, directing and controlling. A
factor analysis reported by Dunnette (1986) of
65 managerial activities yielded seven factors:
monitoring the business environment, planning
and allocating resources, managing individual
performances, instructing subordinates, man-
aging the performance of groups, representing
groups and coordinating groups.
Janz, Colquitt, and Noe (1997) identied
four distinct facets in which a leader could give
the team autonomy: planning (e.g., scheduling
the team’s work), product (e.g., suggesting new
products or services), people (e.g., recruiting
and hiring members) and process-related (e.g.,
specifying the development method a team
should use).
Hertel, Geister, and Konradt (2005) re-
viewed a collection of empirical works on the
management of virtual teams and summarized
the management functions of virtual team lead-
ers at various phases in the team’s lifecycle. They
proposed that virtual team leaders are generally
engaged in such activities as personnel selec-
tion, task design, team initiation, performance
management, training and team development,
and disbanding and re-integration.
These studies categorize leader functions
in different ways and use different labels for
these leader functions. Integrating these differ-
ent categorizations, Table 1 summarizes four
major leader function categories which can be
delegated to virtual team members. These four
areas are the overlapping important leader-
manager functions identied in the mentioned
studies. The rst leader delegation category con-
sists of planning-related team management and
leadership activities that a virtual team leader
Table 1. Four-dimensional leader delegation framework
Delegation Aspects Key Activities
Planning-related
Scheduling the team’s work
Setting the team’s long-term goals
Setting the team’s short-term objectives
Setting the team’s budget
People-related Selecting team members
Removing members from the team
Determining team members’ training needs
Process-related
Assigning work to team members
Selecting the tools team members will use in their work
Determining the team’s operating procedures and work instructions
Control-related
Evaluating the progress of the team’s work
Evaluating team product quality
Determining corrective actions when performance objectives are not met
52 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
can possibly delegate to the team. The second
category consists of people-related team man-
agement and leadership activities that a virtual
team leader can delegate, such as team stafng
and team member training. The third category
consists of process-related team management
and leadership activities (or teamwork process
management). The fourth is control-related
team management and leadership activities.
This category relates to the leader’s functions
and activities that control the work progress
and quality of a virtual team.
To keep the categorization parsimonious,
the key activities in each delegation category
may incorporate more than one of the leader
functions identied in the previous studies.
For example, “determining operating pro-
cedures and work instructions” incorporates
two virtual leader functions in MacKenzie’s
model: standardize methods and decide how to
achieve goals. Some of the manager activities
are not included because they either cannot be
assigned to the team through leader delega-
tion or are already tasks that are done by team
members, for example, suggesting new products
or services.
Delegation in Virtual Software
Teams
The issue of productivity loss is very important
when global software team leaders weigh the
benets and costs of delegation. A software team
is often formed for completing certain tasks
within a limited time period. The short-term
nature of the project requires avoidance of any
productivity loss. In a longitudinal study using
18 student information system development
(ISD) teams, Nicholson et al. (2002) conrmed
the importance of avoiding productivity loss in
distributed teams. The researchers attempted to
identify the characteristics/behaviors of effec-
tive ISD project team leaders/managers. Their
exploratory analysis revealed that face-to-face
(FTF) and virtual ISD team members valued
different ingredients of leadership in different
phases of the ISD project. They found that a
key ingredient for virtual team leadership was
having a realistic leader who managed time ef-
ciently. Global virtual teams involved in ISD
are temporary structures that are very focused
on the development of the information system
application. These teams have tremendous time
constraints. Hence, the role of a virtual project
team leader is to remain practical in terms of
the goals set and the deliverables promised,
and to manage the time allocated to each task
efciently. Therefore, to avoid productivity loss,
global software leaders are likely to delegate
more to competent teams than to incompetent
teams. A competent virtual software team has
the knowledge, attitudes and expertise required
to perform the team tasks and is more likely to
plan well and manage their work efciently
(Faraj & Sambamurthy, 2006). A competent
team can quickly apply their expertise to the
delegated task and reduce productivity loss to
a minimum. In contrast, delegation to incom-
petent teams means that leaders have to spend
additional time giving feedback and monitoring
the execution of delegated tasks to ensure that
performance standards are met. Otherwise,
managerial anxiety over loss of control can
be overwhelming. However, in the distributed
global team environment, close monitoring
and timely feedback is difcult as “manage-
ment by walking around” can not be used as a
strategy (Paré & Dubé, 1999). Due to increased
temporal distances, possible increased cultural
distances and the lean nature of computer-medi-
ated communication in virtual teams (Bell &
Kozlowski, 2002), leaders will need to spend
much more time and effort in coordinating,
monitoring and coaching team followers in the
delegated tasks.
Besides the need to manage time efciently,
the matrix organizational structure of global
software teams is often another reason why
team leaders delegate more to competent team
members. Developing subordinates’ skills and
condence is the biggest reason why leaders del-
egate or consult their followers, especially when
followers’ skill sets are still to be developed
(Yukl & Fu, 1999). The potential growth of the
followers is likely to be the major benet leaders
obtain from delegation to an incompetent team.
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 53
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
However, virtual teams are often designed to
cross geographical and organizational boundar-
ies to allow dispersed organizations to maximize
their expertise without having to physically
relocate individuals. The required expertise
for a given task or project may be dispersed at
multiple locations throughout the organization;
however, a global software team may facilitate
the ‘pooling’ of this talent to provide focused
attention to a particular problem without having
to physically relocate individuals (Kayworth
& Leidner, 2000). Therefore, global software
teams are often dispersed in cities or even
countries and, very commonly, the team follow-
ers do not report to the team leader in a direct
organizational line. Due to the typically short-
term nature of the team and the complexity of
the reporting relationship, global software team
managers are more likely to be evaluated on
their success in achieving project goals, rather
than on their development of team members.
In delegating to incompetent teams, global
software team managers—unlike line manag-
ers who may treat the costs of delegation as an
investment to be redeemed later—are faced with
the cost of sacricing team performance, which
determines the manager’s own promotion and
career growth. As managing a matrix structure
is already challenging for virtual team leaders
(Oertig & Buergi, 2006), the costs associated
with delegating to an incompetent team would
tend to deter virtual team leaders from this man-
agement strategy. Based on these arguments,
the following hypothesis is proposed:
Hypothesis 1: Leader delegation will be
positively correlated with global software team
competence.
The Effects of Team Leader
Delegation
In this section, three variables, motivation, team
exibility and team member satisfaction with
their software team leader are each discussed.
An argument is made as to why each of these
variables is expected to be affected by team
leader delegation and also why each of these
variables is a primary determiner of overall
team performance.
According to Herzberg’s
(1968) motivation theories, recognition fullls
workers’ esteem needs and can signicantly
improve employee’s performance. A competent
virtual team typically expects the team leader
to recognize the team’s competency by delegat-
ing more responsibility. Leader delegation will
then improve the team’s sense of self-worth and
motivate the team to work more effectively.
An empirical study found that the autonomy
of virtual team members in determining work
objectives and methods improved the intrinsic
motivation of the team (Kirkman, Rosen, Tes-
luk, & Gibson, 2004). They investigated the
relationship between team empowerment and
virtual team performance and the moderating
role of the extent of face-to-face interaction
using 35 sales and service virtual teams in a
high-technology organization. They found that
team members are more intrinsically motivated
when they believe they have high performance
capability (competence), high responsibility
and authority to carry out work (delegation),
and a meaningful task that could impact the
organization. Piccoli and Ives (2003) found
that student virtual teams were more motivated
and satised with less behavior control. Also, a
team’s acceptance of a decision is highest when
the decision is made by the group (Curtis et al.,
1988). Therefore, we posit that delegation to
a competent virtual team would increase the
team’s motivation
On the other hand, delegation to less
competent virtual teams will put the team in a
difcult situation. Due to their low competence
level, they need close monitoring and constant
coaching from the leader or other experts, which
is difcult and costly to obtain in dispersed
virtual teams. Kayworth and Leidner’s (2002)
observation of a dichotomy between inexperi-
enced members wanting and expecting strong
directive leadership and team leaders who, faced
with the practical constraints of distance and the
lack of a direct management line, would prefer
members to be self-managing. Such dichotomy
hurts the team members’ motivation and per-
54 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
formance. In addition, the relatively short-term
nature of virtual teams, which are often formed
dynamically to cope with emerging projects or
tasks, also means that the team has less time
to learn on the job. Instead of desiring delega-
tion, less competent virtual teams want detailed
directions from the leader.
Therefore, we posit that delegation to
less competent virtual teams may not improve
team motivation. Based on these arguments,
Hypothesis 2a is put forth:
Hypothesis 2a: Delegation to competent virtual
software teams will improve team motivation
more than delegation to less competent virtual
software teams.
, as with individual
performance, is a function of ability and mo-
tivation (Jarvenpaa et al, 1998). Sridar, Nath,
Paul and Kapur (2007) have shown that team
member motivation and trust affect performance
in student teams distributed between the United
States and India. Signicant improvement in
team performance is therefore expected from
motivated teams. Based on this argument,
Hypothesis 2b is put forth:
Hypothesis 2b: Virtual team leader delegation
indirectly improves team performance through
improving virtual team motivation.
Team flexibility refers to
how exible a virtual team is in responding to
environmental and personnel changes. Most
virtual teams are knowledge-based teams which
are formed to solve customer problems or to
develop new products (Kirkman et al., 2004).
The complex, knowledge-based tasks many
virtual teams perform require behaviors such
as planning and executing, managing team
performance, improving team processes, and
inuencing organization-level direction and
resource allocations (Mohrman, Cohen, &
Mohrman, 1995). In conducting these activi-
ties, teams have to make sense of their tasks,
improvise their work processes, and adjust how
they make progress toward agreed upon goals.
Therefore, exibility is important to virtual team
performance and team leaders should delegate
to competent virtual teams to allow them to
exibly adapt to their immediate situations
and opportunities. Remote team leaders may
not be able to understand the work context or
to appreciate the consequences of the changes
occurring in the distributed locations. Delega-
tion, therefore, puts this task in the hands of
the competent virtual team members. These
members can also make decisions in a more
timely matter than the leader. Having appro-
priate authority delegated to them, they can
proactively inuence team leaders’ decisions or
even their own decisions instead of passively
waiting for managerial permission before tak-
ing actions. Therefore, delegation to competent
virtual teams will increase team exibility. In
contrast, delegation to less competent virtual
teams may not improve team exibility since
less competent team members may not have
the skills to make decisions and form action
plans. Based on this argument, Hypothesis 3a
is put forth:
Hypothesis 3a: Delegation to competent virtual
software teams will improve team exibility
more than delegation to less competent virtual
software teams.
When a virtual team exibly adapts to
its work situations and is free to respond to
situations in at timely manner, it will be more
risk-taking and learn from experience to con-
tinuously improve its work processes and to
perform in more efcient ways. Based on this
argument, Hypothesis 3b is put forth:
Hypothesis 3b: Virtual team leader delegation
indirectly improves team performance through
virtual team exibility.
:
Delegation to competent virtual teams repre-
sents the leaders’ recognition of their compe-
tence. Delegation allows the competent team to
utilize their capabilities to adapt to the immedi-
ate opportunities and changes without waiting
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 55
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
for decisions to be made by the distant leader.
Therefore, delegation to a competent virtual
team should improve the team’s satisfaction
level with the leader.
On the other hand, delegation to a less
competent virtual team may decrease the team’s
satisfaction with the leader. In a global virtual
team study consisting of undergraduates as fol-
lowers and experienced MBA graduate students
as team leaders, Kayworth and Leidner (2002)
found that the inexperienced undergraduate
followers were more satised with leaders who
gave clear, detailed instructions and feedback.
Faraj and Sambamurthy’s (2006) study of 65
software development teams found similar
results. Their study showed that empowering
leadership has an important positive impact on
team performance but only under conditions of
high task uncertainty or high team expertise.
In the software teams they studied, when team
members have signicant levels of professional
experience with software development, they
are more likely to possess the relevant exper-
tise and experience for managing their project
activities. For such teams, an empowering
leadership approach might be more appropri-
ate since members may need less direct control
and coordination and possibly possess as much
relevant task expertise as the team leader. In
contrast, when teams have low professional
experience, a directive leadership might be
more appropriate because the members look
to the leader to provide needed directions and
guidance about their work activities.
As software team distance increases, the
work context exhibits increased complexi-
ties, for example, cultural misunderstandings,
communication difculties, and so forth. This
makes virtual teamwork more daunting for a
less competent team. Under such circumstances,
the team needs to attain condence and a sense
of direction from a strong leader. Based on this
argument, Hypothesis 4a is put forth:
Hypothesis 4a: Delegation to a competent
virtual team will improve the team’s satisfaction
with their team leader more than delegation to
a less competent virtual team.
When the virtual software team members
are satised with the team leader, the team
leader will be more able to inuence the mem-
bers to work towards team goals and therefore
to improve team performance. This has been
conrmed in empirical studies (e.g., Zeffane,
1994).
Based on this argument, Hypothesis 4b
is put forth:
Hypothesis 4b: Virtual team leader delegation
indirectly improves team performance through
improving virtual team’s satisfaction with the
team leader.
RESEARCH STUDY
Study Design and Sample
A full-scale study with industry software devel-
opment teams is currently underway to explore
the presented research hypotheses. We report
here on the pilot study conducted to test the
validity and reliability of the constructs that
were formed for this research and the viability
of our research model. Although many of the
questions that were used came from studies that
had already tested their validity and reliability,
there were modications made to the questions
to (a) t the virtuality nature of the teams being
studied and (b) to t the software development
environment. For example, the constructs of
team competency and leader delegation were
adapted to describe competency in terms of
software skills and portions of leader delegation
in terms of the assignment of software tasks
Student teams were used to pilot the
research survey because the diversity of the
student teams closely matched the software
development team populations that the nal
survey is intended for. The student teams are part
of a computer science and information systems
program at an American, east coast university
that has one of the most diverse student bodies
in the United States. The students are primar-
56 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
ily from China, India, the Middle East, Brazil,
and Pakistan.
Forty-eight students in 30 software-de-
velopment teams took an online survey that
requested information about the variables
presented in the hypotheses. Thirty-two males
and 16 females; three graduate students and
45 undergraduate students participated. All
participants were involved in teams that were
engaged in developing a single software pro-
gram for the entire semester. The software
teams are part of the ABET accredited Capstone
course designed to have students working in
teams on real software projects before they
matriculate. This is a course most students
take in their last year of school. Companies are
solicited for software projects and present their
projects to the class. A team leader volunteers
for a software project and then interviews and
accepts members for his or her team. Teams
normally have four members but some teams
have three or ve members. Teams then meet
with the company representative and develop
requirements, a budget, a design and a deliver-
able tested product. Reports are due, including
a management report with assigned team roles,
at regular intervals. Thus, the teams are set up
to behave as much as possible like typical soft-
ware development teams. Our interviews with
corporate management in software companies
indicates that projects rarely last for more than
six months and that when new teams form, the
membership is also new, so our student teams
represent this type of project assignment.
The research instrument is designed to work
with virtual teams. Thus, it may be asked, how
can this student population constitute a virtual
team? Almost any team at the university studied
is partially virtual because of the nature of the
university. All capstone classes are held in the
evenings because team members usually are
part time students. The university is a com-
muter university with many classes online or
partially online so that students may come to
the university once a month. Thus, many of
the teams meet virtually and much of the team
work is done by e-mail, instant messaging and
teleconferencing. It was felt that with the more
mature students, seniors, the partial virtuality
and member multiculturism presented a popula-
tion suitable for piloting the research.
The survey was given near the end of the
semester so the team members had worked
together for about three months.
In a second administration of the survey,
34 graduate students from 14 report-writing
teams took the online survey. The team task
was to analyze an industry case study and write
a team report based on the case study results.
The teams consisted of ve to six members with
team leaders elected by the team members. The
survey was given after the team nished their
rst case study project. These 14 teams were
taking a two-month summer online management
information systems course.
Only two of these teams reported meeting
face-to-face once a week. The other teams did
not report meeting face-to-face during the team
project and may never have met each other
face-to-face. This second set of teams was also
different from the rst in that they were work-
ing on a report rather than developing software.
Thus, they met the requirement of being virtual
for our pilot study but not of being a software
development team. We use this second team to
determine if the difference in virtuality might
have an inuence.
After the survey, open-ended interviews
with six members from two software-develop-
ment teams were conducted face-to-face. Each
interview was conducted by two researchers and
videotaped with the interviewee’s permission.
The interview results are reported in section
4.4 and 4.5.
Survey Measurement
In this section, each of the constructs used in the
research is described in more detail, in particular,
its source for validity and reliability verica-
tion are cited along with a sample question that
presents the intent of the measure.
The four categories of virtual
team leader delegation were measured by 13
7-point Likert-scale items in the survey. Sec-
tion 3.1 introduced the conceptualization of the
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 57
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
four categories. A sample question statement is
“how much is your team able to schedule team
work?” (completely – not at all)
: In the rst round of
the survey, six questions were used, which
are adapted from the situational leadership
measurement of follower ability (Hersey &
Blanchard, 1988) and Hardin, Fuller, and
Valacich’s instrument of virtual team efcacy
(2006). A sample question is “The team has past
experience related to the team job” (strongly
agree – strongly disagree). In the second round,
six 7-point Likert scale questions were used to
assess specic skills important to team tasks. A
sample question is “how do you evaluate your
team on its critical analysis skills?” (extremely
high – extremely low )
: Four 7-point Likert
scale items measuring this construct are adapted
from situational leadership theory (Hersey
& Blanchard, 1988). A sample questions is
“The team is motivated to take on additional
responsibilities if needed to nish the project”
(strongly agree – strongly disagree).
: Three 7-point Likert-
scale items were created by the research team
consisting of ve researchers with extensive
virtual team research experience to measure
this variable. A sample question is “This
team quickly responds to new opportunities”
(strongly agree – strongly disagree).
:
Three 7-point Likert-scale items were created
by the research team to measure this variable.
A sample question is, “I am dissatised with
the way the team leader manages this project”
(strongly agree – strongly disagree).
Team performance
is measured by ve 7-point Likert-scale items
adapted from Henderson and Lee’s (1996) study.
This is a composite measure which reects
ve important areas of software team work
outcomes including: the amount of work the
team produces, the efciency of team opera-
tions, the team’s adherence to schedules, and
the team’s adherence to budgets, the quality
of work the team produces. This measurement
was examined by ve experienced researchers
and three IT managers with extensive virtual
team experience and was considered by them
to provide a valid and comprehensive view of
software team outcomes. A sample question
of this measurement is “Compared to other
projects you have served on or observed, how
do you evaluate your team’s performance
on adherence to schedules” (Extremely high
– extremely low).
: In addition to
these constructs, the study also captured data
on another two variables: trust towards other
team members, and interdependence of tasks
performed by team members, which were found
to moderate the impact of leader delegation
(Yukl & Fu, 1999) Trust is measured with four
7-point Likert scale items from Jarvenpaa et al.’s
study (1998). A sample question measuring trust
is “If I had my way, I would not let the other
team members have any inuence over issues
that are important to the project” (strongly agree
– strongly disagree). Task interdependence is
measured by two items adapted from Campion,
Medsker, and Higgs (1993). A sample question
is “To what extent do the team members have to
share work materials to get the project done?”
(strongly agree – strongly disagree). Finally,
team background information such as team
member’s age, year in school, how often the
team met and how they met (remotely or face-
to-face) was also gathered. This information
is used in the post-hoc analysis to help us in
our interpretation of the resulting relationships
uncovered in the research model.
Preliminary Data Analysis
: A principle
component analysis (PCA) was conducted to
test if delegation is a four-dimensional variable.
However, in the pilot study, the student teams
did not have budget constraints and were not
allowed to change their membership once the
teams had been established. Therefore, the four
leadership and management function measure-
ments are not included in the data analysis done
with PCA because students answered these
questions as not applicable. PCA results show
58 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
that all the remaining nine items measuring
delegation load on one component instead of
three unique components. This indicates that
virtual team leader delegation in the student
teams is not multi-dimensional counter to our
predictions.
The numbers in
the study were too small to conduct a factor
analysis, but wherever possible, the questions
used for the constructs were drawn from pre-
viously validated surveys. We also checked
the constructs for face validity by reviewing
the questions with twenty experts from the
countries where the virtual teams for the full
study were located. In addition, a card sorting
test was performed on the constructs using 20
respondents. Ninety-ve percent of the ques-
tions were sorted correctly supporting a case for
acceptable construct validity. Finally, external
validity is a concern because students were used
in the study. We treat this issue in the discussion
on this research’s limitations.
: Except for
trust (α=0.409), the Cronbach Alphas of other
construct measurements are above the level of
0.8. Trust is therefore not included in further
data analyses.
: Due to the small
sample size, a simple measure was used to judge
within-team agreement level: individual team
members’ responses were considered to have
an adequate level of within-team agreement
and were averaged to obtain a team score if the
difference between the highest score and the
lowest score in a team was less than 2.5 (half the
scale range). There was a high level of within-
group agreement in more than 85% of the 44
teams on all constructs in the research model.
Therefore, individual team member’s responses
are averaged to get team-level data.
Test of Hypotheses
Multiple regression testing was chosen to
analyze the data as the data met normality and
homogeneity of variance requirements (An
arcsine transformation was carried out on the
team competence and performance measure to
achieve these assumptions). Structured equation
modeling was not used because of the small
sample size and the intent of the study (pilot).
Although a PLS model is more likely to have
more accurate beta scores, it also has a greater
chance of a Type I error (Goodhue, Lewis, &
Thompson, 2006). Since this was a pilot study,
we wanted to bias it against possible spurious
results.
: Hypothesis 1 predicts
that virtual team leaders delegate more to
competent virtual teams than to less competent
virtual teams. A multiple regression analysis
was conducted to test this hypothesis. The test
results shown in Table 2 support Hypothesis 1.
Delegation is positively correlated with team
competence.
Hypotheses 2, 3, and 4 Tests: Hypothesis
2a predicts that leader delegation to competent
virtual teams will improve team motivation
more than delegation to less competent virtual
teams. Hypotheses 3a and 4a predict the effects
of virtual team leader delegation on team ex-
ibility and on a team’s satisfaction with team
leader respectively. The three hypotheses were
tested by stepwise regression with the outcome
variables regressed on delegation, team com-
petency and the interaction term of delegation
and team competency.
Table 3 presents the test results in the
software-development teams. Hypotheses 2a
and 3a are not supported, as no signicant
interaction effects were found. Therefore, the
effects of delegation on team motivation and
team exibility were not found to change as team
competence varied. Regarding Hypothesis 4a,
the results show that leader delegation improves
the team’s satisfaction but such effects do not
change as team competence varies. In addition,
team competence was found to signicantly
improve team motivation, exibility, and team
satisfaction with the leader (p<0.05).
Table 4 shows the Hypotheses 2a, 3a,
and 4a test results in the report-writing teams.
Regarding Hypothesis 2a, Team leader delega-
tion signicantly improves team motivation
(p=0.006). However, the effects of leader
delegation on team motivation did not differ as
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 59
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
Table 3. Hypotheses 2a, 3a, and 4a test results in software-development teams
Motivation Flexibility Satisfaction
Delegation 0.537 0.821 1.502*
Competency 1.536*** 1.423*** 1.708***
Delegation
X Competency 0.11 -0.153 -0.231
R-Square 0.75 0.54 0.65
F-Overall 21.93*** 7.91*** 9.94***
*p < 0.10, ** p < 0.05, ***p < 0.01
Table 4. Hypotheses 2, 3 and 4 test results in report-writing teams
Motivation Flexibility Satisfaction
Delegation 0.782*** 0.216 1.502***
Competency 0.126 0.069 1.708***
Delegation
X Competency -0.433 0.102*** -0.231***
R-Square 0.531 0.687 0.561
F-Overall 12.263*** 21.912*** 11.502***
*p < 0.10, ** p < 0.05, *** p < 0.01
team competence level varied. Hypotheses 3a
and 4a are supported by the regression results.
The interaction effects of delegation and team
competence are signicant such that leader
delegation to competent virtual teams improves
team exibility and team satisfaction with the
leader more than delegation to less competent
virtual teams.
Hypotheses 2b, 3b, and 4b Tests: Hypoth-
eses 2b, 3b, and 4b predict leader delegation
indirectly improves virtual team performance
through improving team motivation, exibility
and satisfaction with leader respectively. A
Sobel test determines the signicance of the
indirect effect of the mediator by testing the
hypothesis of no difference between the total
effect and the direct effect. This method is used
because its superiority of reducing Type 1 error
and increasing power (Mackinnon, Lockwood,
Hoffman et al., 2002). Due to the small sample
size of the study, a bootstrapping Sobel test
was performed.
Table 2. Hypothesis 1 test results in both report writing and software development teams
Delegation Regressed on Team Competency
Software Develop-
ment Teams Report Writing
Teams
Standardized
Coefcient 0.508*** 0.706***
R-Square 0.258 0.498
F-Overall 8.706*** 9.913***
*p < 0.10, ** p < 0.05, *** p < 0.01
60 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
For the ,
the tests did not nd delegation improved
motivation and exibility. Consequently, the
effect of delegation motivation and exibility
improvements on virtual team performance was
not testable and therefore, Hypotheses 2b and
3b are rejected. Hypothesis 4b is supported by
Sobel test results, as shown in Table 5. There-
fore, delegation improves team performance
partly through improving team satisfaction
with the leader.
For the , Hypothesis
2b is not supported by regression test results,
as shown in Figure 2. Regression tests did not
nd that motivation had a signicant positive
impact on performance. Therefore, delegation
was not found to improve team performance
indirectly through improving team motivation.
Previous tests on Hypotheses 3a and 4a did not
nd that delegation improved team exibility
and satisfaction with the leader. Consequently,
delegation cannot be tested as to whether it
improves virtual team performance through
improving team exibility and satisfaction with
leader. Therefore, Hypotheses 3b and 4b are not
supported either.
Interview Data Analysis
The interviews were analyzed in two steps.
First, two PhD students and two professors with
extensive research experiences watched the
videos of the interviews and discussed what they
found relevant to this study in a group meeting.
Each of them explained his/her ndings to the
group and the group discussed the validity and
implications of these ndings. These discussion
results were recorded in writing. Second, one
PhD student summarized these group discus-
sion results and combined them into a set of
statements. Then, she watched the interview
videos for the second time. She put down
quotes from the interviewees related to each
statement, examined these quotes to judge how
they supported or disproved the categories of
ndings and modied the results of the analysis
accordingly. This two-step procedure does not
transcribe the interview data but roughly fol-
lows the process of creating a coding scheme,
coding the data, iteratively improving the coding
scheme and drawing conclusions. This proce-
dure uses insights made by multiple researchers
who separately viewed the interview videos and
therefore should be adequately rigorous for a
small-scale pilot study.
Table 5. Sobel test results on hypothesis 4b in student software-development teams
Sobel
Statistic F Value Percentage of Total Effects
that are Mediated Ratio of the Indirect Effects to
the Direct Effects
2.868 0.004 77.34 3.412
Figure 2. Hypothesis 2b test in report-writing teams
Delegation
Motivation
Performance
0.782***
0.091
0.860*
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 61
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
The interview data analysis found overall
that delegation was an important part of team
leadership and affected team outcomes in vari-
ous ways. The following paragraphs present the
detailed ndings.
First, team leaders were aware of the impor-
tance of delegation, and engaged in delegation
for a variety of reasons. One team leader learned
from his previous military training that leaders
should teach their subordinates to do their job
so that they can take over if the leader is not
available. The other reason both team leaders
gave for delegating to the team was a belief
that the team had the capabilities to perform
the work. This belief in the team’s competence
came from knowledge about team members’
past performance and their professional expe-
riences. Also, one team leader recognized that
competent team members have egos and decided
that being too controlling would hurt such
egos. Lastly, at least one team leader felt that
delegating tasks to incompetent team members
with the added support of fellow team members
could be a way to build up a team member’s
self-condence and skills.
Second, as the team develops, a leader’s
delegation style may change. In both teams the
author interviewed, the leaders’ style changed
from one of controlling to that of delegating as
team member’s competence increased. In one
team, members reported that at the beginning
of the project, the team leader “tried to control
everyone and everything in the project.” When
the leader was traveling, he sent team members
detailed e-mails that specied the work to be
done and the precise methods to use. As his
team members demonstrated their capability
and produced several deliverables ahead of
deadlines, the leader “eased off in the middle.”
Near the end of the project, the team leader was
very trusting and allowed the team members to
self-lead. One of the most competent members
in the team even led the team for two weeks
when the leader was on vacation in a foreign
country.
Third, delegation happened with monitor-
ing and coaching. After certain tasks or functions
were delegated, the leaders monitored how the
team performed and coached the team members
when needed. One team leader, for example,
suggested books for the team members to read
and advised the team on available develop-
ment tools.
Fourth, delegation affects team motivation
and a team’s satisfaction with the leader. Within
one team, a team member missed the rst dead-
line and produced very low-quality deliverables.
The team leader took action and became very
directive setting detailed work schedules for
the team, making detailed work assignments
to team members and closely monitoring the
work quality and progress of the team. Such
non-delegation drove the team members to work
harder. In this case team members appreciated
the team leader’s direction and effort to get their
project started. In the other team, team members
had adequate professional experience and the
skills needed to perform the project tasks. In the
beginning the team leader was very controlling
and the team members complained and even
had a confrontation with the leader. However,
as the team leader became aware that he was
directing a competent team, he switched to
negotiating with team members when making
decisions. With this change, the team members
felt his trust in their abilities and became more
satised with the leader.
DISCUSSION
The following hypotheses were supported by the
study. First, virtual team competence predicted
leader delegation behaviors. This implies that
virtual team leaders should carefully evaluate
a team’s competence before delegating tasks,
especially before delegating important tasks.
Second, leader delegation improved the sat-
isfaction and motivation of team members.
The effects of delegation on exibility and
satisfaction were more prominent in competent
report-writing teams than in less competent re-
port-writing teams but these effects were small.
Other hypothesized results were not found.
None of the intermediate outcome variables
(exibility, motivation, satisfaction) mediated
62 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
the effects of delegation on team performance.
Instead, tests results indicated that delegation
directly improves virtual team performance.
Figure 3 presents the model supported by the
study results.
Comparing the results found in the two
types of teams, one will observes that rst, del-
egation exerted deeper inuence on the report-
writing teams than the software-development
teams. It is suspected that the differences in
delegation effects may arise from the differences
in the number of times the teams met face-to-face
or, as suspected, the degree of virtuality of the
team. In contrast to the software development
teams, which met face-to-face at least once
a week, the report-writing teams barely met.
Students in the report-writing teams were in
an online summer course and, throughout the
project, only two teams met face-to-face once
a week. As collaboration and communication
processes suffer from lack of face-to-face
contact, the leader’s role in team coordination
and communication becomes more important.
Therefore, leader delegation produced deeper
effects in the more virtual report-writing teams.
This suggests that leader delegation is likely
to be a very signicant factor in managing
virtual teams.
Limitations of Study
One of the major limitations of this study is
that the teams were student teams working on
class projects. In particular, some of the ques-
tions that were designed for corporate virtual
teams were not applicable. Student teams do
not typically assign salaries, manage nances or
hire and re personnel. Only the planning part
of the delegation construct showed differences
between teams. The other parts were scored as
not applicable (one of the possible answers) so
that the results were pooled into one construct
called Delegation which consisted mostly of
items that addressed planning delegation.
In addition, the software and report writing
teams did not represent the distributed teams that
the survey was prepared for. There were some
examples of distributed teams, for example,
one-half of the team lived in the southern
part of the state and the other half lived in the
northern part, but, by and large, teams consisted
of members who were individually virtual but
also met face-to-face occasionally. The report
writing teams were the most virtual with some
team members never having seen each other.
It can also be argued that student teams
do not give representative answers that paral-
lel those of individuals working in companies
although there is evidence that this is not
always so. Hughes and Gibson (1991) found
Figure 3. Delegation effects on student teams. (Bold lines indicate supported hypotheses. The
double line indicates a new result showing a direct link between delegation and team perfor-
mance.)
Delegation
Competence
Satisfaction
Flexibility
Motivation
Performance
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 63
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
that MBA students made decisions comparable
to managers in an Executive MBA program,
but Ashton and Kramer (1986) in their lit-
erature review, note that attitude questions
are answered differently by individuals in the
workforce than by students. Briggs, Balthazard,
and Dennis (1995) found students to be valid
predictors of managerial technology adoption
and Remus (1989) found graduate students to
be more representative of industry personnel
than undergraduates. In particular, studies show
that students are not representative because of
their lack of experience in the workforce and
because of their youth. Because most of the
students in our study were part time students
and had full-time jobs, because the age of the
students in the teams was higher than normal
for university undergraduates and because the
students represented the cultural mix that we
wanted to assess, it was felt that this study’s
population was representative of the industry
group our study targeted.
We did nd, however, evidence that sug-
gests that students were responding differently
than a workforce population might. Student
groups uniformly evaluated their team compe-
tence and team performance highly with more
spread on this evaluation in the report-writing
teams (Master’s students). We also found that
we were not able to obtain any viable reliability
on what is considered a highly reliable trust
measure that we borrowed from the literature.
We therefore did not include trust in our models.
We also remade the competence measure into a
formative construct for the survey of the report
writing team. This helped to x the skew in the
distribution of these results.
As mentioned earlier, the skewed evalua-
tions of team competence and performance may
be an artifact of student teams, but they also
might be an artifact of virtual teams in some
cultures. Thus, an additional variable to collect
and compare to self report of team performance
is a team leader’s report of team performance
in addition to other related variables such as
subproject completion times. We continue to
strive to obtain measures of performance from
team leaders, but, to date, the descriptions of
how management scores performance for their
distributed teams lead us to believe that self-re-
port of perceived performance is not any worse
a measure than managerial reports. In particular,
the cultural and temporal distances are likely to
affect a leader’s perception of performance.
The trust measure may have been highly
unreliable because student team members have
a different social relationship than workforce
team members. One trust question asked if a
team member would like to control the work
of the other team members. This is socially out
of the question in a student team where fellow
team members are independent individuals
more than members of some greater whole
such as a company. Thus, team members are
likely to respond negatively to this question.
A second question asked if a team member
felt uncomfortable with the work of the other
team members. A team member could answer
positively in a socially acceptable way to this
question. Thus, the reliability of a construct that
works ne in a company environment falls apart
in the student context. However, this failure of
the trust construct could be applicable to virtual
teams where their non-colocated counterpart
may also feel that they have no right to con-
trol or monitor the work that the distant team
performs even if they feel uncomfortable with
this lack of control. This is likely to be true
of teams in China and India where new hires
are continually being added to the workforce.
None of these younger team members would
feel that they should control the work of their
team members in Europe or North America,
but their recent training may also make them
feel uncomfortable with the work being done
in these places.
These discussion notes that the literature
demonstrates that student teams can provide reli-
able answers that represent industry situations if
the groups are appropriately chosen, but it also
suggests that the very nature of student teams
might be more appropriate for studying virtual
teams across cultures in that their responses
might represent similar cultural responses.
The small sample size also limits the gener-
alizability of the study ndings. We analyzed the
64 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
teams separately because of inherent differences
in their virtuality and work assignments. We
also performed a separate analyses so that we
could examine the effects of virtuality although
confounded with task and a more senior student
population. This made the sample sizes small.
Approximately 200 software teams were asked
to ll in the survey but only 30 responded. The
response rate was extremely low, in particular,
because many of the students were in their
nal year, already had jobs that reduced the
importance of the payment incentive we of-
fered and were quite busy with class projects.
The response rate was signicantly higher for
the report writing teams (about 50 percent) but
the class size was small. We also choose to ana-
lyze all of the teams, even those with only one
respondent because the response rate was low.
Thus, there were 12 software teams that only
had one member. For the report writing teams,
only teams with two or more respondents were
used in the analysis. The problem with a single
team member responding constitutes another
analysis problem because that single member
could have been an outlier generating data
unrepresentative of the team. Studies now in
progress with a larger number of industry teams
will yield more conclusive ndings related to
virtual team leader delegation.
CONTRIBUTIONS AND
FUTURE WORK
This study addressed an important yet under-
researched area regarding leader delegation
practice in virtual software teams. It provides
statistically sound conclusions which reduce the
confusion arising from the conicting ndings
of prior case studies. Some virtual software
team managers claim that the increased distance
make self-management necessary while others
question whether excessive autonomy will
produce negative effects in a complex virtual
team environment. A pressing question for
industry practitioners is when and how much
authority and responsibility should be delegated
to remote team members. Our ndings suggest
that delegation is an important virtual team
management strategy that positively affects
team performance. In particular, the results from
the student teams imply that a team leader will
delegate in response to his or her sense of how
competent the team is. However, the real world
situation affects this delegation with a push and
pull effect, that is, management will want to
delegate more because the task of managing a
global virtual team means more communication,
odd hours of work scheduled for communication
and care needed to avoid miscommunications.
This is the push to delegate. However, the pull
effect is that a team leader because of the lack of
information on the global virtual team stemming
from language difculties, cultural differences
and simply not being able to observe team
behavior because of the distance will not want
to delegate to the team. The ndings suggest
that team leaders need to be trained to ignore
these effects and perform their delegation based
on real information about the team, perhaps by
visiting the virtual team or setting up viable
measures for team performance.
The tendency of the student teams to give
self-reports of high competence and high per-
formance suggests leadership guidelines for
industry. In particular, it would be wise to give
team leaders training in the cultures they are
interacting with so that they can better judge
the self-reports they are obtaining.
Our ndings also suggest that delegation
is a good thing in that it increases a team’s
satisfaction with its leader, a team’s exibility
to adjust the project to local needs and a team’s
motivation. The ndings, however did not nd
a strong mediating relationship between these
values and team performance Earlier discussion
on the limitations of the study suggest that the
performance measures captured were cor-
rupted by the use of student teams. Flexibility,
motivation, and satisfaction have been shown
to affect performance in face-to-face teams so
there is good reason to believe that obtaining
better measures of performance would nd
moderation by these variables. This is future
work that needs to be performed.
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 65
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
Overall, more delegation was found to be
a positive behavior for a team leader, but with
the youth and newness of team members join-
ing virtual teams in many of the companies that
offshore or outsource, this is likely to be a poor
strategy unless measures are taken to bring up
the skill sets of the offshore team members.
Due to high team member turnover in
some countries, especially with the constantly
increasing wages in these countries, companies
are reluctant to invest in training for these team
members. However, as we have been told by
team managers from India, this training is pre-
cisely one of the mechanisms used to reduce
turnover. This is another variable that clearly
needs evaluating in future work.
Overall, the pilot study ndings suggest
useful recommendations for virtual software
team leadership as to when and how much they
should delegate to the team based on the team’s
degree of virtuality and competence. However,
the hypotheses of this study were mainly based
on software team literature and the pilot study
was done mainly with student software teams.
Future work with real industry teams with
varied types of companies and a variety of
organizational models needs to be performed
to verify these ndings. Currently a full-scale
survey and interview has been conducted with
the global software testing teams of a Fortune
100 company. Future data analysis will be per-
formed to compare the industry team ndings
with the ndings of this pilot study.
REFERENCES
Ashton, R. A., & Kramer, S. S., (1980). Students
as surrogates in behavioral accounting research:
Some evidence. Journal of Accounting Research,
18(1), 1-15.
Bass, B. M. (1990). Bass and stogdill’s handbook of
leadership (3rd ed.). New York: The Free Press.
Bell, B.S., & Kozlowski, S.W.J. (2002). A typol-
ogy of virtual teams: Implications for effective
leadership. Group and Organization Management,
27(1), 14-49.
Briggs, R. O., Balthazard, P. A., & Dennis, A. (1995).
Graduate business students as surrogates for execu-
tives in the evaluation of technology. Journal of End
User Computing, 8(4), 11-17.
Campion, M. A., Medsker, G. J., & Higgs, A. C.
(1993). Relations between work group character-
istics and effectiveness: Implications for designing
effective work groups. Personnel Psychology, 46,
823–850.
Cofn, T. E. (1944). A three-component theory of
leadership. Journal of Abnormal and Social Psychol-
ogy, 39(2), 63-83.
Curtis, B., Krasner, H., & Iscoe, N. A., (1988). Field
study of the software design process for large systems.
Communications of the ACM, 31(11), 1268-1287.
Davis, R. C. (1942). The fundamentals of top man-
agement. New York: Harper.
Eveland, J.D., & Bikson, T.K. (1988). Work group
structures and computer support: A eld experi-
ment. Transactions on Ofce Information Systems,
6(4), 354-379.
Faraj, S,. & Sambamurthy, V. (2006). Leadership of
information systems development projects. IEEE
Transactions on Engineering Management, 53(2),
238-240
Goodhue, D., Lewis, W., & Thompson, R. (2006).
PLS, small sample size, and statistical power in MIS
research. Proceedings of the 39th Hawaii International
Conference on System Science, (pp. 1-10).
Hardin, A., Fuller, M., & Valacich, J. (2006). Measur-
ing group efcacy in virtual teams: New questions in
an old debate. Small Group Research, 37, 65-85.
Henderson, J. C., & Lee. S. (1996). Managing IS
design teams: A control theories perspective. Man-
agement Science, 6, 757-777.
Hersey, P., & Blanchard, K. (1988). Management of
organizational behavior: Utilizing human resources
(5th ed.). Englewood Cliffs, NJ: Prentice-Hall.
Hertel, G., Geister, S., & Konradt, U. (2005). Man-
aging virtual teams: A review of current empirical
research. Human Resource Management Review,
15(1), 69-95.
Herzberg, F. I. (1968). One More Time: How do You
Motivate Employees? Harvard Business Review.
January-February, 109-120.
66 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
Hughes, C. T., & Gibson, M. L. (1991). Students
as surrogates for managers in a decision-making
environment: An experimental study. Journal of
Management Information Systems, 8(2), 153-166.
Janz, B. D., Colquitt, J. A., & Noe, R. A. (1997).
Knowledge worker team effectiveness: The role
of autonomy, interdependence, team development,
contextual support variables. Personnel Psychology,
50(4), 877-905.
Jarvenpaa, S. L., Knoll, K., & Leidner. D. (1998).
Is anybody out there? Antecedents of trust in global
virtual teams. Journal of Management Information
Systems, 14(4), 29-64.
Jarvenpaa, S.L., & Leidner, D.E. (1999). Communi-
cation and trust in global virtual teams. Organization
Science, 10(6), 791-815.
Kayworth T. K., & Leidner D.E. (2000). The global
virtual manager: A prescription for success,. Euro-
pean Management Journal, 18(2), 183-194.
Kayworth, T. K., & Leidner. D. E. (2002). Leader-
ship effectiveness in global virtual teams. Journal of
Management Information Systems, 18(3), 7-40.
Kirkman, B. L., Rosen, B, Tesluk, P. E., & Gibson,
C. B. (2004). The impact of team empowerment on
virtual team performance: The moderating role of
face-to-face interaction. Academy of Management
Journal, 47(2), 175-192.
MacKenzie, R. A. (1969). The management process
in 3-D. Harvard Business Review, 47(6), 80-87.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J.
M., West, S. G., & Sheets, V. (2002). A comparison of
methods to test mediation and other intervening vari-
able effects. Psychological Methods, 7, 83-104.
Mohrman, S. A., Cohen, S. G., & Mohrman, A.
M. (1995). Designing team-based organizations:
New forms for knowledge work. San Francisco:
Jossey-Bass.
Oertig, L., & Buergi, T. (2006). The challenges of
managing cross-cultural virtual project teams. Team
Performance Management, 12(1), 23-30.
Paré, G., & Dubé, L. (1999). Virtual teams: An
exploratory study of key challenges and strategies.
Proceedings of the 20th International Conference on
Information Systems. Charlotte, NC: Association for
Information Systems.
Piccoli, G., & Ives., B. (2003). Trust and the unin-
tended effects of behavior control in virtual teams.
MIS Quarterly, 27(3), 365-395.
Remus, W. (1989). Using students as subjects in
experiments on decision support systems. In 22nd
Hawaii International Conference on System Sciences,
HICSS’22. IEEE Computer Society Press.
Sridar, V., Nath, D., Paul, R., & Kapur. K., (2007,
July 25-27). Analyzing factors that affect perfor-
mance of global virtual teams. Proceedings of the
2nd International Conference on Globally Disributed
Work (pp. 159-169). Bangalore, India.
Townsend, A.M., DeMarie, S.M., & Hendrickson,
A.R. (1998). Virtual teams: Technology and the
Workplace of the Future. The Academy of Manage-
ment Executive, 12(3), 17- 29.
Urwick, L. F. (1952). Notes of the theory of or-
ganization. New York: American Management
Association.
Yukl, G. (2002). Leadership in organizations (5th
ed.). Upper Saddle River, NJ: Prentice Hall.
Yukl, G., & Fu, P. P. (1999). Determinants of del-
egation and consultation by managers. Journal of
Organizational Behavior, 20(2), 219-232.
Zeffane, R. M. (1994). Correlates of job satisfaction
and their implications for work redesign: A focus on
the australian telecommunications industry. Public
Personnel Management, 23(1), 61-76.
Zhang, S., Tremaine, M., Fjermestad, J., Milewski,
A., & O’Sullivan, P. (2006), Delegation in virtual
teams: Moderating effects of team maturity and
team distance. In proceedings of IEEE International
Conference of Global Software Engineering (pp.
62-68).
International Journal of e-Collaboration, 5(1), 47-68, January-March 2009 67
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
Suling Zhang is an assistant professor in the graduate management studies department of Kean University.
She received a PhD of information systems from New Jersey Institute of Technology, an MS in information
technologies from University of North Texas and a BA in english literature from China University of Law
and Political Science with minor in law. Her research interests are virtual team leadership, communication
technologies in distributed teams and small business technology management. She has published in journals
such as International Journal of e-Collaboration and Journal of Enterprise Infomration Management.
Marilyn Tremaine is a research professor at Rutgers University where she has joint appointments in the
College of Communication, Information and Library Science and the Department of Electrical and Computer
Engineering. Prior to her appointment at Rutgers, she chaired the Department of Information Systems at
the New Jersey Institute of Technology. Tremaine has also been a professor in the Department of Computer
Science at the University of Toronto and the Michigan Business School, spending ten years in each of these
positions. Her current research focuses on the design and development of multimodal interfaces to be
used in collaborative settings, on developing computer-based training that enhances spatial skills for the
comprehension of complex visualizations, on examining the impact of leadership styles, social networks,
and cultural differences virtual team software development
Allen Milewski is currently an associate professor of Software Engineering at Monmouth University in
West Long Branch, NJ, USA. His research interests include team collaboration, global teamwork and
cultural inuences on social and cognitive processes - specically information-seeking strategies and the
effects of distributed team composition within software development. He received his PhD in experimental
cognitive psychology and was an associate professor of psychology before moving to the software industry.
While working in industry, Milewski accrued more than twenty years of practical experience as a usability
practitioner, researcher and inventor with ten patents. Since moving from industry back to academics, he
has studied and published in the areas of internationalization of user interface, visualization, and human
attention. Most recently, he has been funded to research and design collaborative information systems for
homeland security and emergency management use.
Rich Egan is a senior university lecturer of information systems at the New Jersey Institute of Technology.
Rich has successfully defended his dissertation for a doctorate in information systems, and has an MBA in
MIS from St. Peters College. He is also a certied information systems auditor. His research interests include
cultural variation and its impact on global software development, business continuity, and the impact of
regulatory efforts on software development. Before entering academia Rich spent over 28 years in the IT
industry providing him with experience in all aspects of information systems and technology.
Patrick O’Sullivan is a “system test” architect at IBM’s Dublin Software Lab. He completed his BSc in
1991, his MSc in 1997 and his PhD in 2001, and both of his IBM sponsored MSc and PhD HCI research
projects won industry prizes. Pat has over 16 years software engineering experience with the latter 10 years
at IBM. His research and publications interests include HCI, software engineering, software process/ar-
chitectural models, software globalization, UIMS, RAS, performance engineering and test automation. Pat
is a senior member of the IEEE, a chartered engineer with the Institute of Engineers of Ireland and also
holds the FEANI title of EurIng. He was a founding and committee member of the SoftTest Ireland SIG
and is an external examiner with the University of Limerick, Ireland. Pat is also a member of the Institute
of Engineers of Ireland’s Information and Communications Technology Board.
Jerry Fjermestad is an associate professor in the School of Management at NJIT. He received his BA in
chemistry from Pacic Lutheran University, an MS in operations research from Polytechnic University,
an MBA in operations management from Iona College and an MBA and PhD from Rutgers University in
management information systems. Jerry has taught courses on management information systems, deci-
sion support systems, systems analysis and design, electronic commerce, data warehousing, and graduate
seminars in information systems. His current research interests are in collaborative technology, decision
support systems, data warehousing, electronic commerce, global information systems, customer relation-
ship management, and enterprise information systems. He has co-edited eleven special issues in various
68 International Journal of e-Collaboration, 5(1), 47-68, January-March 2009
Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global
is prohibited.
IS journals. Jerry’s publications include one edited book on CRM and over a hundred scholarly papers in
a number of journals and proceedings. He serves on the editorial board for several journals and organi-
zations, including Journal of Information Science and Technology, the International Journal of Electronic
Collaboration, and the International Journal of Information Security and Privacy.
Reproducedwithpermissionof thecopyright owner.Furtherreproductionprohibitedwithoutpermission.
