Citation
The role of social capital in determining BPO outcomes

Material Information

Title:
The role of social capital in determining BPO outcomes
Creator:
Ghosh, Biswadip
Publication Date:
Language:
English
Physical Description:
xi, 105 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Infrastructure (Economics) ( lcsh )
Information technology -- Management ( lcsh )
Information resources management ( lcsh )
Contracting out ( lcsh )
Offshore outsourcing ( lcsh )
Knowledge management ( lcsh )
Contracting out ( fast )
Information resources management ( fast )
Information technology -- Management ( fast )
Infrastructure (Economics) ( fast )
Knowledge management ( fast )
Offshore outsourcing ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 96-105.).
General Note:
Department of Computer Science and Engineering
Statement of Responsibility:
by Biswadip Ghosh.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
176888546 ( OCLC )
ocn176888546
Classification:
LD1193.E52 2007d G46 ( lcc )

Downloads

This item has the following downloads:


Full Text
THE ROLE OF SOCIAL CAPITAL IN DETERMINING BPO OUTCOMES
by
Biswadip Ghosh
B.Tech., Indian Institute of Technology, 1986
M.S., Colorado State University, 1988
M.B.A., Regis University, 1993
A thesis submitted to the
University of Colorado at Denver and Health Sciences Center
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Computer Science and Information Systems
2007


This thesis for the Doctor of Philosophy
degree by
Biswadip Ghosh
has been approved
by
Ellen Stevens
Date


Ghosh, Biswadip (Ph.D., Computer Science and Information Systems)
The Role of Social Capital in Determining BPO Outcomes
Thesis directed by Assistant Professor Judy E. Scott
ABSTRACT
Organizations are pursuing the outsourcing of business processes (BPO) to offshore
locations. However, current research has shown that knowledge management issues
between the client and vendor organizations leads to less than expected benefits in many
BPO arrangements. This research applies the Resource Based View (RBV) framework
to study the impact of Social Capital resources on building Knowledge Management
Capabilities in the BPO as well as the ultimate BPO outcome. The essence of RBV
theory is that when investments are made into building IS resources, the impact of those
investments on the firm's performance manifest through the firms capabilities.
Complementarity theory also suggests that a resource produces greater returns in the
presence of another resource or capability than it can produce by itself. Firms pursuing
BPO may be at different stages of the deployment of knowledge management
capabilities and the IS resources of infrastructure and social capital and therefore the
capabilities of the engagement may not be being fully exploited. Hence the impact of
social capital and infrastructure resources on knowledge management capabilities
and/or their combined impact on BPO outcome could explain the variability seen in BPO
outcomes. A measure of BPO outcome based on the three information flow costs from
coordination theory vulnerability, production and coordination is also developed. A field
study of a knowledge management system (KMS) in a BPO is used to test the
hypotheses. Results indicate that increasing social capital resources alone and as well
as in combination with increasing infrastructure resources in the client-vendor BPO


infrastructure resources in the client-vendor BPO arrangement leads to an increase in
knowledge management capabilities. However, increasing infrastructure resources alone
has no significant effect on building knowledge management capabilities. Moreover, the
increase in the knowledge management capabilities also significantly increased BPO
outcome. Finally, this study also confirmed that IS resources can intervene to strengthen
the relationship between building knowledge management capabilities and business
process outsourcing outcomes as a small complementarity effect was supported. This has
an important implication for practitioners, who seek to improve BPO outcomes that
capabilities in the BPO arrangement must be developed in conjunction with the deployment
of IS resources.
This abstract accurately represents the content of the candidates thesis. I recommend its
publication.


DEDICATION
I dedicate this thesis to my parents, Jaya and Bireswar Ghosh, who gave me an
appreciation of learning and taught me the value of perseverance and resolve. I further
dedicate this to my wife, Dalia, for her unfaltering support and considerable
understanding while I was completing this work. I also dedicate this thesis to my two
children, Juhi and Avishek for their acceptance of the many hours I spent in my home
office, often forgetting them, as they stood in front of the French doors of the study trying
to get my attention and a moment of my time.


ACKNOWLEDGMENT
My thanks to my advisor, Dr. Judy E. Scott, for her contribution and support to my
research. I wish to thank Dr. Jahangir Karimi for his considerable input and guidance. I
also wish to thank all the members of my committee for their valuable participation,
review, recommendations and insights into this work and its final form.


TABLE OF CONTENTS
Figures...............................................................x
Tables...............................................................xi
CHAPTER
1. INTRODUCTION.......................................................1
Resource Based View............................................3
Information Systems Resources..................................4
IT Infrastructure Resources..................................5
Social Capital or Relationship Resources.....................6
Knowledge Resources..........................................8
Knowledge Management Capabilities..............................8
Research Questions............................................14
2. OUTSOURCING RESEARCH BACKGROUND...................................15
Outsourcing Research..........................................15
Outsourcing Relationship Research.............................17
Outsourcing Outcomes Research.................................19
3. THEORETICAL FOUNDATIONS...........................................21
Social Capital Theory.........................................22
BPO Outcomes..................................................30
Coordination Theory...........................................31
4. DEFINING STUDY CONSTRUCTS.........................................33
Social Capital Resources......................................33
Group Identification........................................34
Generalized Trust...........................................34
vii


Pro Sharing Norms............................................35
Infrastructure Resources.......................................36
Knowledge Management Capabilities..............................36
BPO Outcomes...................................................37
5. RESEARCH HYPOTHESES................................................41
6. RESEARCH METHODOLOGY...............................................50
Research Design................................................50
Study Variables................................................50
Measurement of Research Variables..............................51
7. DATA COLLECTION....................................................53
Collection Methodology.........................................53
Organizational Practices for Knowledge Management..............53
Outsourced Processes...........................................55
Customer Service Process.....................................55
Network Consulting Process...................................56
Survey Results.................................................57
Knowledge Resources............................................61
8. DATA ANALYSIS AND RESULTS..........................................62
Estimation of Internal Consistency.............................68
Dimensionality, Convergent and Discriminant Validity...........70
Second Order Model.............................................72
Hypothesis Testing.............................................73
Comparing Moderator Model with Second Order Model..............76
9. DISCUSSION AND CONCLUSIONS.........................................80
Implications for Research......................................82
Implications for Practice......................................83
viii


Generalizability of the Results...........................84
Limitations...............................................85
APPENDIX
A. FINAL SURVEY.................................................88
B. OUTER MODEL LOADINGS (DIRECT EFFECTS MODEL)...............90
C. OUTER MODEL LOADINGS (SECOND ORDER MODEL).................93
BIBLIOGRAPHY .........................................................96
ix


LIST OF FIGURES
Figure
1.1 Research Background.......................................................13
3.1 Theoretical Foundations...................................................29
5.1 Research Model............................................................41
8.1 Direct Effects Model......................................................63
8.2 Second Order Model........................................................64
x


LIST OF TABLES
Table
4.1 Survey Instruments............................................................39
7.1 Demographics across Processes.................................................59
7.2 Descriptive Statistics of Items...............................................60
8.1 Correlations, CR and AVE for Direct Effects Model.............................66
8.2 Correlations, CR and AVE for Second Order Model...............................67
8.3 Internal Consistency Measures.................................................69
8.4 Structural Model Results......................................................71
8.5 Correlation Matrix of Factors in ISR Second Order Model.......................72
8.6 Principal Components Analysis for Second Order Model for ISR..................75
8.7 Component Score Coefficient Matrix for ISR....................................75
XI


CHAPTER 1
INTRODUCTION
The concept of transferring the operational ownership and execution of one or
more business processes is referred to as Business Process Outsourcing (BPO). In an
offshore BPO, the vendor organization is located in a different country from the client
organization, that is seeking to outsource. Due to the rapid globalization of information
technology and improvements in telecommunications networks, firms are pursuing
offshore outsourcing to remote countries like India and China. Offshore outsourcing
allows the domestic firms to tap into large pools of educated workers at significantly lower
labor costs. Additionally, client firms benefit by getting access to large markets for
products and services in the foreign countries and take advantage of time zone
differences to staff round the clock operations using a follow the sun" policy (Lieberman,
2004). The vendor organization undertakes responsibilities of the execution of the
process. Examples of BPO are seen in several industries such as finance, healthcare
and technical support. Both core and supporting (non core) processes in a client firm's
value chain are being outsourced to offshore vendors.
The offshore BPO market is the fastest growing segment of the overall IT outsourcing
market and is projected to grow 60% year over year (Trapper, 2003). However, risks in
offshore outsourcing are significant and can lead to less than expected benefits (Aron,
Clemons and Reddi, 2005; Scott and Ghosh, 2004). Actual outsourcing case studies
indicate that many organizational goals remain unfulfilled in these arrangements (Lacity
and Willcocks, 1998). Their study reported that only 54% of the outsourcing agreements
that they surveyed realized cost savings. These numbers were recently confirmed by a
practitioner study by (Moore, 2004; Moore and Martorelli, 2004). Such variability in the
1


BPO outcomes call for empirical research into the drivers for success in such BPO
arrangements.
Current BPO research literature points out the need for bi-directional knowledge
management capabilities in offshore BPO arrangements. While the risks and challenges
are significant in an offshore outsourcing relationship, knowledge management practices
have been found to play a major role to improve the effectiveness of these arrangements
(Willcocks, Lacity and Kern, 1999). Using case studies, Rottman and Lacity (2004)
identified 20 best practices that can be used to overcome the difficulties. While this
study did not specifically study KM practices, yet KM initiatives such as balanced
scorecard metrics, real time dashboards and identifying subject matter experts were part
of the list of 20. A study of knowledge management practices in IT outsourcing by
Willcocks, et. al. (2004) indicated the importance of social capital resources in
outsourcing relationships.
Another case study of the organizational impact of the outsourcing decision at
Logistics Information Systems Agency (LISA) presented by Willcocks, Lacity and Kern
(1999) found several risks that are not fully considered when the BPO was implemented,
but later on needed management resources to alleviate. These issues are pronounced in
the high technology industry, where knowledge based processes are being outsourced.
High technology firms differ from other organizations in that they operate in a more
dynamic and innovative environment with the need for efficient information flow. Existing
research studies on outsourcing and associated organizational factors have to date
ignored the specific characteristics of the BPO for such knowledge based processes
(Lacity and Willcocks, 1995, Levina and Ross, 2003, Carmel and Agrawal, 2002). An
important question that has not been studied in IS research is how relational and
2


infrastructure resources contribute to the BPO outcome for knowledge intensive
processes in the high technology industry. As more and more knowledge intensive work
is moved offshore, the need for bi-directional knowledge management capabilities in the
BPO arrangement is increasing particularly for professional and high technology work
processes (Ghosh and Scott, 2005d; Ghosh and Scott, 2006a).
Resource Based View
The Resource Based View (RBV) of the firm states that resources are essential
raw materials for capability-building and their availability determines the firms ability to
build such capabilities, which are often critical drivers of firm performance (Wade and
Hulland, 2004; Barney, Wright and Ketchen, 2001; Bharadwaj, 2001). Capabilities are
defined as repeatable patterns of actions in the use of those resources to create, produce
and offer products/services to the market. Capabilities can include management ability
and skills and processes/systems that allow for creation, storing and sharing of
knowledge (Wade and Hulland, 2004). Resources can be independently valued and
traded, while capabilities cannot. The RBV framework supports modeling complementary
relationships between resources and capabilities to determine whether they have a
complementary impact on outcomes (Ray, Muhanna and Barney, 2005). The outcomes
are defined as the unit of the firms performance. Complementarity arises when a
resource produces greater returns in the presence of another resource than it produces
by itself. While the resource-based view (RBV) of the firm recognizes the
complementary role of resources, it is not well developed in the business process
outsourcing research and theory. However, studies in inter organizational alliances and
partnerships have applied RBV and found that it is a suitable theoretical foundation to
explain the outcome of the alliance and partnerships (Lavie, 2006). BPO falls in the
3


domain of strategic networks (Gulati, Nohria and Zaheer, 2000) and therefore, RBV is a
suitable framework to apply to study these inter organizational arrangements.
Information Systems Resources
One of the benefits of applying RBV theory is that several prior studies had
categorized and defined Information systems resources, which provides a foundation for
incorporating applicable IS resources into this study. The RBV literature provides several
different categorization of Information Systems resources that have been used in RBV
research. Ross, Beath and Goodhue (1996) identified technical, human, and relationship
resources as three categories of IS resource. Another categorization used by Bharadwaj
(2000) consists of tangible, human, and intangible resources. The common theme
among the above categorizations fall into two resource areas tangible infrastructure
resources and intangible human relationship management resources. Wade and Hulland
(2004) listed 8 categories of Information Systems resources (1) Managing external
relationships, (2) Market Responsiveness, (3) Managing internal relationships, (4) IS
planning and change management, (5) IS Infrastructure, (6) IS technical skills, (7) IS
development, and (8) Cost Effective IS Operations. They also recommend that for
research, which examines specific technologies, models using a set of more narrowly
defined resources is appropriate. Feeney and Willcocks (1998) framed the outsourcing
question on IS resources and recommend that non core IS resources be outsourced by
the firm. They identify 9 IS resources (1) Business Systems Thinking, (2) Relationship
building, (3) Leadership, (4) Contract Facilitation, (5) Vendor Development, (6) Contract
Monitoring, (7) Architecture Planning, (8) Informed buying and (9) Making Technology
work. Arguably the first six resources can be seen as relationship and people based,
while the last 3 account for the physical infrastructure.
4


As seen from the previous section, two broad classes of IS resources need be
considered for this study- (1) relationship resources and (2) infrastructure resources.
These resources are heterogeneously distributed across the two firms in the BPO and
impact operations on either side, and their presence (or absence) and complementarities
with knowledge management capabilities may explain the variability of BPO outcomes.
IT Infrastructure Resources
Infrastructure resources refer to a firms shared IT assets (e.g., hardware,
software tools, and networks, databases and data centers). They are the foundations for
a firms IT architecture, which is the blueprint to support multiple business processes and
user groups across the two firms in the BPO arrangement. Reliable IT infrastructure
resources will ensure the success of IT architecture, which tends to be highly firm specific
and evolves over a long period of time during which gradual enhancements are made to
reflect changing business needs (Karimi, Somers and Bhattacharjee, 2007a). As the
infrastructure becomes distributed throughout firms and even beyond their boundaries in
a BPO, the firms need a clear vision of where to locate individual technology components
and responsibility for those components. Firms with a valuable technology asset are
developing architectures that elaborate rules for distributing hardware, software, and
support as well as controlling their accessibility to user groups based in multiple firms -
independent of individual applications. These rules specify what kinds of data to share
and how to store them, where to locate servers, and how to support applications and
technologies (Feeney and Willcocks, 1996). Firms without well-defined architectures
have dealt with the challenges of distributed computing by first delivering systems and
then thinking about how to connect and support access to them. The result is that their
systems are either poorly supported or expensive to operate or both. In firms with a well-
5


defined architecture, infrastructure integration requirements and support considerations
drive system design so that new systems not only meet business needs but are also cost
effective.
Social Capital or Relationship Resources
Social capital (relationship) resources consists of sharing ownership, risk and
responsibility of the firms business operations across the client and vendor personnel
(Ross, Beath and Goodhue, 1996). Ross, Beath and Goodhue (1996) found that close
working relationships across firms allow staff to observe business processes in action
and accumulate experience in solving real business problems. Enabling interactions is
the central foundation of social capital theory (SCT). Nahapiet and Ghoshal (1998) define
social capital resource as 'the sum of the actual and potential resources embedded
within, available through, and derived from the network of relationships possessed by an
individual or social unit. Social capital thus comprises both the network and the assets
that may be mobilized through that network.
SCT theory argues that effective knowledge combination, exchange and transfer
occur when sufficient opportunities for interactions and exchange are present in the inter-
organizational relationship between individual knowledge workers (Nahapiet and
Ghoshal, 1998). Social capital is defined in three dimensions with each dimension
contributing to the meaning of social capital and where each dimension alone is not able
to fully capture the concept in its entirety. The three components are commonly seen as:
(1) Generalized Trust or the relational aspects along with types/tones of social
interactions, (2) Identification or the issues that relate to structural identification with the
cross firm work group, and (3) Pro Sharing Norms or the cognitive issues of rules and
norms governing social action (Coleman, 1988, Putnam, 1993, Nahapiet and Ghoshal,
6


1998). Social capital resources only exist within a relationship and cannot be separated
from the network. They are created through exchanges that happen when the parties in
the relationship facilitate opportunities for interaction (Ghosh and Scott, 2007b). In a
large empirical study of public organizations in Singapore, components of social capital
such as organizational norms and structure were found to moderate the usage of
electronic knowledge repositories in managing knowledge (Kankanhalli, Tan and Wei,
2005). Given these results, it is clear that social capital resources in the relationship
between the client and vendor personnel is a key concept in BPO arrangements that
need to be explored further.
Social capital resources form an important complementary resource to the
infrastructure resources, which have traditionally been deployed to manage outsourcing
projects in the IT and IS domains (Rottman and Lacity, 2004; DeLone, et. al., 2005).
Such social capital resources have the potential to mitigate risks and address
unaccounted for scenarios that are not specified in the traditional work contracts. Social
capital resources have also been found to be an important component in the successful
adoption and use of knowledge management systems (Wasko and Faraj, 2005;
Huysman and Wulf, 2006). The research on inter-organizational knowledge
management also shows that knowledge management between organizations is affected
by four factors that include the characteristics of the organizations and their relationship
(Argote, 1999). Ko, Kirsch and King (2005) reported that the more arduous the
relationship, the less knowledge transferred, while greater the shared understanding,
credibility and motivation, greater the knowledge transferred. Inkpen and Tsang (2005)
also found that social factors play a significant role in knowledge management in social
networks. The stream of research in partnerships and inter-organizational learning has
7


also identified that social capital in relationships play a role in knowledge transfer in both
formal and informal settings. Burges (2005) found both individual relational resources
and organizational infrastructure resources impacted knowledge management across
firms. Similar results in the domain of inter-organizational relationships can be noted
from Scott (1998), Malhotra, Gossain and El Sawy (2005) and Argote (1999).
Knowledge Resources
A third category of IS resources identified in RBV literature are the knowledge
resources (Karimi, Somers and Bhattacharjee, 2007a). Knowledge resources are the
knowledge assets that are in the knowledge management system. These knowledge
assets or resources are the artifacts of the BPO arrangement. These knowledge artifacts
fall into three categories product knowledge, customer knowledge and managerial
knowledge (Tanriverdi and Venkatraman, 2005). Product knowledge refers to research
and development and operations knowledge by which the firm develops and produces its
products and services. Customer knowledge refers to needs, preferences and
intelligence about customers and markets of the firm, specifically in the areas of
marketing and advertising skills and policies. Managerial knowledge refers to the
knowledge required for governing the business units of the firm and include managerial
practices, policies and processes of the firm (Grant, 1991).
Knowledge Management Capabilities
Cross unit knowledge management capabilities are important sources of synergies
in multi-firm arrangements such as BPO (Tanriverdi, 2005). These capabilities require
coordination processes across the firms as well as knowledge assets or artifacts. Since
knowledge has a personalization aspect to it, hence it is more difficult to manage and
manipulate than mere information. Knowledge management capabilities are defined as
8


organizational processes that allow the firms to create, exDloit and protect these
knowledge assets (Alavi and Liedner, 2001; Davenport and Prusak 1998). Gold,
Malhotra and Segars (2001) identified a set of knowledge management processes that
an organization utilizes to transfer and exploit knowledge from external sources, such as
partnerships. They mentioned that these processes require the presence of social capital
resources and suitable IT infrastructure resources deployed in the firm. They identified
four dimensions acquisition, conversion, application and protection.
Zahra and George (2002) identified these processes as a reconceptualization of
the absorptive capacity of a firm, as dynamic capabilities, which four underlying
organizational processes which are each comprised of specific routines and procedures
that allow the identification of knowledge and its exploitation. They list absorptive
capacity in knowledge management as either (1) potential capacity, which consists of
acquisition and assimilation capabilities or (2) realized capacity", which are the
transformation and exploitation capabilities. These capabilities can exist across firms by
the institution of a set of organizational processes that lets the firms coordinate
knowledge management activities. There are five interrelated processes that are critical
for managing cross-unit knowledge synergy: (1) Creation or Acquisition of Knowledge
(Nonaka, 1994; Gold, Malhotra and Segars, 2001), (2) Integration or Conversion of
knowledge ((Grant, 1996), (3) Leverage or application of knowledge (Gold, Malhotra and
Segars, 2001) and (4) Transfer of knowledge (Argote, 2001, Szulanski, 1996). A final
component of knowledge management capability is the (5) protection of knowledge
(Gold, Malhotra and Segars, 2001), which is particularly important in offshore BPO
arrangements, where intellectual capital abuse is quite common and well documented in
9


practitioner literature. This research stream also suggests that if any one of these
capabilities is low, then firm performance can be reduced (Tanriverdi, 2005).
Hence, knowledge management capabilities are a set of capabilities that provide a
reflection of the firms overall capacity to exploit knowledge in its business. The current
research notes the need for knowledge resources that are relevant and applicable across
multiple business units is essential for cross unit knowledge synergies. Those resources
need to be transferred and integrated with business operations for the firm to realize
value. Finally the resources must be leveraged by changing the behavior of the firm in
order to achieve the expected performance benefits. These knowledge capabilities
provide cross unit synergies and mutually support each other (Venkatraman and
Tanriverdi, 2005).
Research specifically aimed at studying the impact of social capital resources on
the outcome of offshore BPO relationships, while being well motivated from the IS
literature, is still lacking. Such social capital resources are seen to impact knowledge
management capabilities in prior research, while knowledge management capabilities are
seen to improve inter organizational outcomes. Hence, a study to model the impact of
social capital resource on the knowledge management capabilities in a BPO and the
ultimate outcome of a BPO holds promise. Such a study has not been attempted in the
IS research literature, and holds significant promise to explain the ambivalent results of
BPO (Lacity and Wilcocks, 1998).
Additionally, researchers of SCT have called on research into the diffusion and
exploitation of the social capital resources in downstream knowledge management
capabilities in the organization, as well as measuring the impact of SCT using transaction
and process costs (Nahapiet and Ghoshal, 1998). While social theories can be used to
10


ascertain benefits in inter-organizational knowledge management, information flow based
cost measures are needed to get a better unbiased measure of outcomes. Such a
measurement of BPO outcome using information costs is also lacking in the IS research
stream. Coordination theory is based on the coordination, production and information
costs associated with the interactions of organizational subunits (Malone and Crowston,
1999). A measurement model based on coordination theory can shed light on the
effective flow of information and knowledge and the allocation of resources, both of which
are critical components of any business process (Malone and Crowston, 1994) and
therefore are critical to a BPO. Hence, using coordination theory to evaluate the outcome
of offshore BPO holds promise.
This research posits that high technology firms that invest in IS resources -
relational and infrastructure to facilitate knowledge management capabilities to support
the BPO may see improved BPO outcomes through the impact of those investments in
resources on the capabilities in the inter organizational partnership. Indeed outsourcing
success is significantly linked to factors that fall beyond the written contract between the
client and vendor (Koh, Ang and Straub, 2004). These include IT infrastructure
resources and the development of relationship/social capital in the form of human
resources within the outsourcing teams. Complex business obligations can never be
completely written up. Koh, Ang and Straub (2004) stress the need to have
psychological contracts that can drive the behavior of vendor personnel. The focus is
on the individual level rather than at the organizational level. This individual focus is
consistent with the working of knowledge processes in professional organizations such
as technical support and healthcare (Ghosh and Scott, 2005b, Ghosh and Scott, 2005c,
Ghosh and Scott, 2006b). Often the intent of complex BPO contracts are lost among the
11


vendor employees and leads to over specification. Improved BPO success can be
achieved through relationship building at the individual level to support the established IS
infrastructure resources that have already been deployed (Hislop, 2002). However,
neither the impact of social capital on knowledge management capabilities in a BPO nor
its impact on BPO outcome have been studied in IS literature. This research poses the
following fundamental question:
Can Information Systems (IS) resources consisting of: (1) Social Capital resources
from social capital theory and (2) Infrastructure resources facilitate knowledge
management capabilities in a BPO?
This study postulates that greater IS resources in the BPO relationship can directly
lead to improved knowledge management capabilities, which results in improved BPO
outcomes. Moreover, the increased IS resources in combination with the increased
knowledge management capabilities also results in improved BPO outcomes, beyond
that produced by the capabilities alone.
In summary, from a theoretical perspective, this study aims to investigate the role
of social capital and infrastructure resources in offshore BPO arrangements by
measuring the impact of those IS resources through its facilitation of knowledge
management capabilities using the resource based view (RBV) of the firm. This study is
motivated by the BPO and social capital literature and holds promise of explaining the
variability in observed BPO results, as well as providing a validated model to measure the
BPO outcome in future BPO research. The Figure 1.1 summarizes the above research
literature review in this conceptual domain.
12


Figure 1.1 Research Background
13


Research Questions
The goals of this research project are to answer the following questions using the
RBV framework to model the deployment of IS resources to support inter firm capabilities
in an offshore BPO arrangement:
1. Can IS resources particularly social capital (from social capital theory)
resources translate to improved knowledge management capabilities
between the client and vendor in the offshore BPO arrangement?
2. Do knowledge management capabilities positively impact the BPO outcome?
3. Do IS resources and knowledge management capabilities have a
complementary relationship with BPO outcome?
14


CHAPTER 2
OUTSOURCING RESEARCH BACKGROUND
Outsourcing Research
The research stream in outsourcing shows considerable diversity in four
dimensions (1) the stage of the BPO ranging from planning to implementation, (2) the
organizational level where the study was conducted, (3) the theory applied and (4) the
methods used. IS researchers have investigated different stages in the outsourcing
decision process starting with the initial outsourcing decision to the outcome of the
decision and have applied various theories and methods at different levels of analysis
from the industry to individual workers (Dibbern, et. al., 2004). The staged decision
making model separates the process of making the decision to outsource into four stages
- (1) why outsource, (2) what to outsource, (3) how to manage the outsourcing
implementation, and (4) finally measuring the outcomes of outsourcing.
Several studies have investigated the questions of why a firm decides to
outsource and what processes or systems they choose to include in the outsourcing
arrangements (Loh and Venkatraman, 1992; Lacity and Willcocks, 1995; Hu, Saunders
and Gebelt, 1997). These studies have found that the determinants of the outsourcing
decision comes from both inside and outside the firm. The impetus for the decision to
outsource is primarily strategic. Factors such as the firms desire to reduce costs and
their need to increase operational focus have been found to be statistically significant in
this decision stage (Ang and Straub, 1998). Likewise, external industry trends and
influences have been found to be significant as well in the decision to outsource (Loh and
Venkatraman, 1992). These external influences include the so-called Kodak effect,
where the firm's decision to outsource is significantly influenced by the decision of other
15


firms in their industry as well as industry best practices and edicts from industry
consultants.
The content and selection of the processes to outsource, however, has been
found to be a decision primarily driven by internal aspects of the firm and with lesser
influence of industry and external factors. Both strategic and tactical characteristics of
the firm have been found to be significant in the choice of processes to outsource (Lacity
and Willcocks, 1995; Lacity and Willcocks, 1998; Ang and Slaughter, 1998). The choice
can be a strategic one taken at the firm level, however it has influences from the tactical
level in terms of political, structural and cultural characteristics of the firm as well as the
profile of the processes themselves.
The nature of outsourcing is defined by the use of a single or multiple vendors,
while the degree of outsourcing is defined as the percentage of the function being
outsourced up to complete outsourcing. Both of these characteristics can be determined
by the strategy and culture of the firm(s) under study as well as the best practices in the
firms industry. Additionally, the type of outsourcing general versus transactional versus
functional (business process outsourcing) is also a determinant of the why and what
decisions (Hu, Saunders and Gebelt, 1997).
The above research papers studies have utilized various theories including
transaction cost theory, game theory, resource theory and other strategic management
theories, and various relationship theories including social exchange theories. The
selection of the theory to use is coupled with the level of analysis, such that resource
based theories have been used at the organizational level to study internal determinants
of the firms resources and capabilities (Barney, 1991), while relationship based theories
16


have been utilized to study cooperation and interactions at the individual level (Kern,
1997).
Outsourcing Relationship Research
The second group of outsourcing studies have focused on the implementation
aspects of the outsourcing decisions to delve into the question of how. These studies
have investigated issues in the selection of the vendor and the building, structuring and
management of the outsourcing relationship between the client and vendor (Kern, 1997;
Grover, Cheon and Teng, 1996). Industry and firm level factors have been identified to
be significant in the selection of the outsourcing vendor. While written contracts and
service level agreements are important monitoring aspects of the outsourcing
relationship, firms have recognized that psychological relationship management is an
important part of managing the ongoing outsourcing contract (Koh, Ang and Straub,
2004). Various firm level characteristics have been found to have a significant influence
in the outsourcing relationship trust, cooperation, social and cultural bonds, shared
vision, expectations and norms as well as commitment (Kern, 1997; Willcocks and Kern,
1998; Grover, Cheon and Teng, 1996).
Outsourcing arrangements come in varied time duration short to long term; and
contractual arrangements from a loose 'pay for service arrangement to stronger
partnerships such as joint ventures. Studies have also found relationships form a link
between how an outsourcing project is structured and how it is managed, however, the
length of the association has no direct impact on the relationship between client and
vendor. For more strategic partnerships or alliances, the need for flexible governance
mechanisms become more important. In such scenarios, the overall conclusion has
been that contracts are necessary, but not sufficient for outsourcing success, rather
17


relationship building becomes more important as the outsourcing becomes increasingly
strategic for the firm (Clark, Zmud and McGray, 1995). The authors have defined and
used several constructs to measure relationships to include such constructs as trust,
communications that include adhoc channels, cooperation, etc. (Grover, Cheon and
Teng, 1996).
Various social and relational exchange theories and contract theories have been
utilized in these studies to investigate process and management issues in outsourcing
relationships between the client and vendor. Case studies and other empirical
approaches using both positivist variance based models as well as interpretist descriptive
methods have been used in this stream of research. Kern (1997) built a model that
extended the structural contract theory models to include relational determinants. These
studies placed their level of analysis at the worker level, which was found to be
appropriate to evaluate relationships between client and vendor (Willcocks and Kern,
1998). Other studies have been done at the individual level and have focused on the
work processes of outsourced IS professionals and the nature of the process/function
being outsourced (Schultze and Boland, 2000). They found that the nature of work
practices can jeopardize the relationship aspects in the outsourcing when the relationship
is asymmetrical with weak group identification and vendors are setup as fall guys for
potential organizational failures. In another study at the individual/functional level,
Sabherwal (1999) found that outsourced projects proceed through cycles that involve
trust, structure and performance. Moreover, when a balance is established between
contractual structure and relationships the client firm is better able to manage changes
and resolve unanticipated problems through BPO management.
18


Thus it is evident from the above that studies that focus on vendor selection and
contractual provisions are well suited at the firm level, while investigations into the
relationship issues in a BPO can be more illustrative at the functional or individual level.
Outsourcing Outcomes Research
A final group of outsourcing research has focused on evaluating the outcome of
outsourcing. These papers have studied the outcomes or success using constructs that
are based on the realization of the expectations of the outsourcing from the firms
perspective (Lee and Kim, 1999; Aubert, et. al., 1999). Measures used in the studies
include such constructs as cost savings, satisfaction, risk avoidance and relationship
quality, such as conflicts at the firm level. Other authors have defined and used
relationship constructs at the individual level such as fulfillment of responsibilities,
treatment of workers, performance differentials and trustworthiness in the relationship
(Ang and Slaughter, 1998).
Various theories have been used to develop measures of BPO outcome.
Transaction cost theories (TCT) were used in a seminal study of outsourcing outcomes
by Lacity and Willcocks (1995). While TCT can be utilized to evaluate certain types of
outsourcing contracts that are deemed classical with limited emphasis on relationships, it
breaks down in the evaluation of outsourcing arrangements in which relationships require
investments that can moderate the outcome. Organizations are not mere substitutes for
structuring efficient transactions as they are established to govern certain kinds of
economic activities through a logic that is very different from that of a market. TCT fails to
recognize this difference, particularly when investments are made into developing
organizational capabilities (Ghoshal and Moran, 1996). Additionally measures based on
behavioral and relationship theories have also been developed (Sabherwal, 1999; Cross,
19


1995; Huber, 1993). A measure of satisfaction was developed based on the Theory of
Reasoned Action and used in a study by Heckmann and King (1994), where they
identified positive and negative indicators of satisfaction using factor analysis.
The major drawbacks of these studies is that outcome or success has been
narrowly measured based on how the firm has defined the meaning of success, that is,
were that firm's expectations realized. This has the problem of having made the
construct value laden. Moreover, these constructs of outsourcing outcome vary from
organization to organization, with limited universal applicability. Hence it is difficult to
utilize these measures in a more generic fashion across different outsourcing scenarios.
Moreover the measurement constructs of outcome are not independent and hence can
have adverse impacts on the measurement model during data analysis. For example
costs reduction can be closely related to reduced service levels. Also it is seen that
these measures are at the firm level and are potentially not suitable for a study at the
functional or individual level of analysis.
20


CHAPTER 3
THEORETICAL FOUNDATIONS
As evident from the previous chapter, a growing concern among organizations,
who are involved in outsourcing, is the post-deployment management of BPO or the
how (Kern and Willcocks, 2000). The focus of this research project is into the how of
post deployment BPO arrangements using the Resource Based View (RBV) of
strategic management. RBV provides a research framework to select IS resources that
are deployed in a domain of study and model their impacts on specific firm capabilities
and performance outcomes. RBV has been successfully applied to multi-firm studies
where the impact of inter organizational collaboration on value creation was studied (Lin,
2006). Several IS research papers discuss the relevance of studying the relational and
infrastructure resources of BPO arrangements together (Koh, Ang and Straub, 2004).
Information Technology is a facilitator of outsourcing through the concept of Net
Enabled Organizations (Zahra and George, 2002). IT Infrastructure capabilities such as
Voice over IP (VoIP) and networking, which makes cross-continental telecommunications
inexpensive, data management systems like storage networks, distributed data
databases and distributed application deployment/deployment platforms such as J2EE
based servers like WebLogic, Websphere, etc. allow firms to pursue offshore outsourcing
strategies very effectively. BPO is often coupled with the introduction of a large IT system
such as an enterprise system like ERP or CRM (Broadbent, Weill and St. Clair, 2003). In
fact the model of BPO is often termed synonymously in the literature as ITES or
Information Technology Enabled Services. This concept of a technology driven
outsourcing strategy is furthered by Sambamurthy, Bharadwaj and Grover (2003), who
describe the phenomena as building dynamic capabilities and flexibility in the firms
21


processes and resources to allow the firm to react to changes in the environment.
Clearly net enablement and IT infrastructure has created opportunities for the
organization to exploit new strategies and react to new opportunities and tap into cheaper
overseas labor markets.
While studying the relevance of the inter-organizational infrastructure deployed in
the BPO is apparent, there are also many justifications for studying the relationships
among the client and vendor personnel. They include the following arguments (1) the
processes and systems do not operate in a vacuum, knowledge workers are important
actors in the model and (2) there are two organizations involved in a BPO the client and
the vendor, each with their own policies, their social systems and characteristics. Hence,
several behavioral theories have been applied to the study of BPO relationships in
studies at the organizational level, workgroup level or even at the individual employee
level. Lyytinen, Mathiassen and Ropponen (1998) studied the risks involved in
outsourcing and how the systems change the shape and behavior of the organizations
and vice-versa. They found a significant reason for the failure of project implementations
is due to the neglect of the relational goals of the organization as well as the social
evolution of the organization, which is often overlooked in the outsourcing models.
Social Capital Theory
Social Capital Theory (SCT) has been proposed in many different terms in
organizational research literature. Ghoshal and Moran (1996) describe social capital as
the organizational advantage". In summary, social capital theory is concerned with the
significance of relationships as a resource for individual action. SCT is often contrasted
with transaction cost theory that is based on human opportunism, while SCT is based on
personal relationships developed through collective norms.
22


Social capital comprises both the network and the assets that may be mobilized
through that network. The social capital resource has three dimensions (1) Group
Identification, which is the structural or identification dimension, which refers to the
network of interactions and how it builds group identity, (2) Generalized Trust or the
relational dimension, which refers to the history of interactions among the people and
how it influences their trust behavior and (3) Pro Sharing norms, which refers to those
resources that provide shared meaning among parties.
The RBV literature also recognizes the components of social capital resource as
important IS resources (Feeney and Willcocks, 1998). They identify Pro sharing norms
as the IS resource of Relationship building, which allows common language and
conventions to build up so that diverse people can interact and work together. They state
that relationship buildings most important contribution is in developing harmony of
purpose and successful communications among staff', which resonates with the earlier
definition of Pro sharing Norms. Another major component of IS resources is the trust
that is developed between business units through a history of interactions, which can
influence their knowledge sharing behavior (Nahapiet and Ghoshal, 1998). Recent
research has shown that trust is an important resource for inter firm knowledge sharing
(Jarvenpa and Staples, 2001). Social capital theory identifies Generalized Trust as an
important relational dimension. Because the development of trust takes time, it can
constitute a component of social capital resource and competitive advantage for the firm
(Karimi, Somers and Bhattacharjee, 2007a).
The final dimension of social capital resource to be considered in this study is
Group Identification, which closely follows the definition of Partnership and Alignment
resources described by Ross, Beath and Goodhue (1996). They describe the case of a
23


large financial services firm that faced intense cost pressures, which they managed to
overcome through strong business partnerships in many divisions. Such team alignments
create group identification that allows a strong focus on the firm's priorities and
outcomes. Thus, Group Identification is a strong IS relationship resource that needs to be
considered in the domain of inter-organizational arrangements such as BPO.
Social capital theory states that the social capital resource is owned jointly by the
parties in a BPO relationship, and no one player has, or is capable of having, exclusive
ownership rights. Moreover, although it has value in use, social capital resources cannot
be traded easily. Social capital resources, therefore, are difficult to imitate because they
develop over many years of cooperation sharing, identification and trust building. Further,
relationship resources, such as social capital resources tend to be socially complex,
relatively immobile, non-substitutable, inimitable, and rare (Wade and Hulland, 2004).
Kankanhalli, Tan and Wei (2005) has applied SCT to model the contextual
factors that help facilitate knowledge contributions to electronic knowledge repositories
(EKR). They found that social capital theory can be applied to explain the behavior of
resources embedded in inter-organizational environments such as networks of practice,
where norms are formed to facilitate knowledge transfer. The benefits of applying social
capital theory to a IS study stem from the ability to model and measure the creation of
structural norms together with the development of relationship aspects such as
generalized trust and group identification on other IS capabilities such as knowledge
(Ghosh and Scott, 2007b).
Of the other social and behavioral theories, the social exchange and
organizational learning theories have been applied to the study of outsourcing as well
(Kern and Willcocks, 2000). Social exchange theory (SET) is based on social psychology
24


and sociology and the theory argues that a series of interactions generate obligations
between people. Hence future action by a person is seen as being interdependent and
contingent on the actions of another person. These interdependent transactions can also
have the potential to generate high quality relationships under certain conditions
(Cropanzano and Mitchell, 2005). IS researchers have utilized SET to study
organizational knowledge transfer scenarios (Kankanhalli, Tan and Wei, 2005; Chen and
Choi, 2005). The authors found that social and individual cost and benefit factors in
knowledge management can be accounted for in SET. The application of SET is suitable
when the participants and contributors of knowledge work under an assumption of a
relatively longer term relationship of interest (Kankanhalli, Tan and Wei, 2005). Thus, a
request for knowledge by one person, if met by the collaboration group at one point in
time results in building an obligation for future reciprocation, supporting a link between
creating social bonds and knowledge management in inter-organizational groups.
Various organizational learning theories have been extended to study inter and
intra organizational learning (Huber, 1991). The concept of organizational learning is
also tied to organizational memory or the ability of the organization to remember norms
and structures from its past experience whether good or bad. Organizational Memory
is considered distinct from individual memory just as organizational learning is different
from individual learning. These concepts are integral to organizational change, as these
processes of learning and memory are what are used by the firm to meet challenges from
the environment and internal forces and enhances its performance. The manifestations
of these processes are often found in the decision making, innovation and information
acquisition and distribution functions (Cohen and Leventhal, 1990; Andreu and Ciborra,
1999).
25


The Task-technology fit literature also shows support for the link between both
social capital resource and infrastructure resource on the sharing of knowledge and
organizational performance (Becerra-Fernandez and Sabherwal, 2001). The theory of
task-technology fit asserts that for a deployed infrastructure to have a positive impact on
individual performance the technology must be utilized and the technology must be a
good fit with the tasks it supports. Their study shows that task inter-dependence and
presenting opportunities for interaction through the infrastructure have a large role in
knowledge combination and management. Therefore, this leads to the notion that if
peoples tasks, goals and objectives are more closely related, such that interactions are
fostered, then it will lead to better combinations of the infrastructure resource and
creation of relational norms and greater knowledge management will occur (Wasko and
Faraj, 2000).
Other studies have reported the benefits of relationship resources on inter firm
cooperation and operations. Kumar, Dissel and Bielli (1998) pointed out the example of
the textile merchants in Prato, Italy. They show how years of face-to-face operations and
close business relationships have built up inherent social capital among a group of textile
merchants in Prato, Italy and shaped their business practices over the years. This
gradual evolution and in-bred relationships have allowed the textile industry in Prato,
Italy to survive multiple shocks the period of fascism, two world wars, the shortage of
raw materials and the emergence of cheap textile producing counties. Regardless of the
environmental and market gyrations, the textile industry in Prato continues to thrive due
to the great business and personal relationships and developed social norms, which has
resulted in improved organizational performance among the merchants.
26


The stream of research on virtual teams also supports the importance of
relationship resources on inter organizational knowledge processes that span multiple
locations. The BPO scenario typically puts non-collocated teams in two separate
organizations the client and the vendor to implement a business process. Research on
virtual teams and the deployment and use of collaborative technologies have also
investigated the inter-organizational phenomena of non-collocated teams building shared
context and exchanging knowledge (Malhotra, et. al., 2001). This research supports the
link between social capital resources and organizational performance. Moreover, these
studies also investigate the inter-organizational dimension and raises the issue of an
environment without common norms for knowledge management. Virtual team research
has investigated the need to create a shared understanding among members before
effective team practices can be established (Jarvenpaa and Leidner, 1999). Results
indicate that such norms and knowledge management take place when planned
opportunities for interactions are made between virtual team members. The development
of a common language allowed better use of collaboration technologies in a case study of
a virtual team at Boeing (Malhotra, et. al., 2001). The virtual environment also was
found to facilitate more knowledge management than a regular collocated group when
working norms are established. These findings are all consistent with knowledge
management in the BPO scenario, where social and structural norms are formed. Their
results indicate that such norms are established and knowledge management take place
when planned opportunities for interactions are made between virtual team members.
These studies note that the development of social capital requires the active and willing
engagement of individuals within an infrastructure. The organization is better placed to
create social capital through accessible infrastructure resource components. Wasko and
27


Faraj (2005) have used SCT to evaluate participation and quality of knowledge
contributions to an electronic network of practice. Their study found that infrastructure
factors and organizational factors can impact the establishment of norms in the inter-
organizational relationships.
IS researchers have also studied factors that improve BPO success rates and
found that better management practices and selective outsourcing can improve
outsourcing success (Lacity and Willcocks, 1998). Management practices are also seen
to help foster relationships and improved outsourcing outcomes (Rottman and Lacity,
2004). Carmel and Agarwal (2002) identify several factors that can facilitate the
alignment of teams across the client and the vendor. Support can also be drawn from the
situated organizational learning literature which places emphasis on the context of
leaning through the creation of structural norms (Lave and Wenger, 1991). This
argument supports the relationship between establishing social capital and norms on
knowledge management. Situated learning theory also claims that individuals can build
shared context and understanding through simultaneous participation in group processes
(Nidumolu, Subramani and Aldrich, 2001). They stress that using this perspective to
study knowledge management provides emphasis on knowledge flows as opposed to
merely knowledge stocks or the artifacts of knowledge management technologies.
Researchers have also utilized situated learning to model and understand the situated
nature of cognition, when structural norms are built as the network actors make sense of
and act coherently in their world (Yuan and McKelvey, 2002).
The above studies imply a clear link between the Social capital resources and
Infrastructure resources on knowledge management capabilities and the combined
impact of those two resources (termed as IS resources) and capabilities on BPO
28


outcome. However, neither of these links have been tested in a BPO scenario in the
existing IS research literature using the RBV framework.
The theoretical foundations are summarized in Figure 3.1.
Figure 3.1 Theoretical Foundations
29


BPO Outcomes
The Information systems research literature shows prior studies that have
investigated outsourcing outcome using various theories Institutional theory (Teo, Wei
and Benbasat, 2003), process theory (Carmel and Agarwal, 2002) and economic theories
(Ang and Straub, 1998; Levina and Ross, 2003). Economic theory suggested that
outsourcing relationships are successful in the cases where the client firm develops a
complementary set of core competencies from the relationship and highlighted the need
to focus on the information processing requirements for the organization, which needs to
be matched with the level of uncertainty facing the organization in the BPO. This
supports the need to measure BPO outcome using an information flow based measure.
Carmel and Agarwal (2002) used process theory to understand the key elements
for outsourcing success at each of the four stages of the clients offshore outsourcing
strategy offshore bystander, offshore experimenter, proactive cost focus using
outsourcing of non-core work and proactive strategic focus with outsourcing of core work
processes. The authors reported the important role of the organizational profile in the
strategic decision making process related to outsourcing by the client firm. Other studies
have investigated the cost advantages that arise from the economies of scale and scope
in the outsourced process possessed by the outsourcing vendor organization (Ang and
Straub, 1998).
Teo, Wei and Benbasat (2003) used institutional theory to understand the factors
that enable the successful adoption of inter-organizational systems. They found that once
the outsourcing arrangements are made, the coercive and normative pressures of
cooperating between the client and the vendor firms will significantly influence the
structure and cultural aspects of both firms. Therefore, we can clearly see that while the
30


initial decision to outsource may be heavily dependent on cost-benefit analysis and to
save operational costs, the outcome is significantly weighed towards the operational
culture and working environments between the two firms -the client and vendor.
Clearly economic theories can be used to make the outsourcing decision,
however the success of the outsourcing decision often ends up in the ongoing process
management of the BPO relationships which the economic theories tend to ignore
(Ghosh and Scott, 2005a; Ghosh and Scott, 2006a, Ghosh and Scott, 2007b). However,
strictly process based measures that do not account for the flow of
knowledge/information in the outsourcing relationship, may not fully explain the observed
outcome.
A measurement using information flow can provide for a broader measure of the
BPO outcome. Without the measurement of any vulnerability of the client firm from lack
of adequate information flow, a true assessment of the BPO outcome is not possible
(Ghosh and Scott, 2006a; Ghosh and Scott, 2005a; Ghosh and Scott, 2004). It is highly
possible that client firms that place greater emphasis on establishing relationship capital
reduce their vulnerability and may fall in the 46% that experience successful BPO
outcomes (Ghosh and Scott, 2005a).
Coordination Theory
Coordination theory, which is based on the coordination, production and
vulnerability costs associated with the interactions of organizational subunits (Malone and
Crowston, 1999) has been used to study the impact of information technology on
organizational structure (Malone and Crowston, 1999). A model based on coordination
theory can shed light on the effective flow of information and knowledge and the
allocation of resources, both of which are critical components of offshore BPO (Malone
31


and Crowston, 1994). Coordination theory offers a vehicle to study the impact of
technology on organizations through the medium of business processes referred to in
the theory as coordination processes. It is well suited to measure the costs of
information/knowledge transfer arising in a BPO relationship (Ghosh and Scott, 2006a;
Ghosh and Scott, 2004; Ghosh and Scott, 2005a). The advantage of coordination theory
is that it allows the study of the organizational impacts of outsourcing at the micro-level -
down in the trenches" using the coordination structures (Malone, 1987). A coordination
structure is defined as a pattern of decision-making and communication among a set of
actors who perform tasks in order to achieve goals. There are three kinds of costs for
these coordination structures production costs, coordination costs, and vulnerability
costs (Tushman and Nadler, 1978). Clearly, all three costs are impacted by outsourcing.
The definition of each of the three costs are:
1. Production costs include the costs of running the outsourced
business processes and will measure any efficiencies or deficiencies introduced.
2. Coordination costs are the costs to manage the communication
between the client and the vendor firms.
3. Vulnerability costs are the unavoidable costs of a changed
situation that are incurred before the organization can adopt to a new situation.
These three costs form coordination theory form three dimensions of the theory
and cumulatively reflect the overall process outcome (Malone, 1987).
32


CHAPTER 4
DEFINING STUDY CONSTRUCTS
Social Capital Resources
The concept of social capital exists with the functional group and should be
measured at the functional level as the creation and semantics of the construct are very
much group or network related (Brewer, 2003). Social capital is multi-dimensional with
each dimension contributing to the meaning of social capital where each dimension alone
is not able to capture fully the concept in its entirety. The main dimensions are commonly
seen as: (1) Trust or relational aspects together with types of social interactions, (2)
Structural issues that relate to Identification with the group, and (3) Cognitive issues that
include rules and norms governing social action (Coleman, 1988, Putnam, 1993).
Nahapiet and Ghoshal (1998) also define three similar dimensions for social capital (1)
structural (referrals, timing, context, network ties), (2) cognitive (shared codes, language
and narratives) and (3) relational (generalized trust, norms, obligations and identification).
Researchers applying SCT have utilized these dimensions in operationalizing social
capital in prior studies (Adler and Kwon, 2002, Kankanhalli, Tan and Wei, 2005). In IS
literature, Kankanhalli, Tan and Wei (2005) have operationalized the social capital
construct using three dimensions (1) Generalized trust, (2) Identification and (3) Pro-
Sharing Norms. These dimensions match the three dimensions in Nahapiet and
Ghoshals (1998) definition of social capital (1) Relational, (2) Structural and (3)
Cognitive. In the inter-organizational context, emphasis is placed on network
configurations, shared cognitive norms and trust (Inkpen and Tsang, 2005). BPO closely
resembles network alliances that involve inter-member social ties as a foundation for
exchange and sharing of knowledge (Gulati, Nohria and Zaheer, 2000). Hence the
33


social capital resource construct must be measured to capture the concept as a reflection
of three interacting dimensions (1) Group Identification, which is the structural or
identification dimension, which refers to the network of interactions and how it builds
group identification, (2) Generalized Trust or the relational dimension, which refers to the
history of interactions among the people and how it influences their trust behavior and (3)
Pro Sharing norms, which refers to those resources that provide shared meaning among
parties (Nahapietand Ghoshal, 1998).
Group Identification
Group Identification is a condition where the interests of the individuals merge
with the interests of the organization, resulting in the creation of an identity based on
those interests. Identification sets the context with which communication and knowledge
exchange occur among organizational members (Nahapiet and Ghoshal, 1998). Three
components of identification that have been used in IS research are (1) sharing of
common values and (2) common goals and (3) strong membership towards the group
(Kankanhalli, Tan and Wei, 2005). When Identification is strong, the effort required for
knowledge contributors may not be a deterrent to knowledge contributions and would
therefore be a facilitator for knowledge management capabilities in a BPO arrangement.
Identification is an extremely important component of the social capital resource in
offshore BPO arrangements due to the natural remoteness and lack of the usual
organizational camaraderie in such scenarios.
Generalized Trust
Generalized trust is an impersonal form of trust that does not rest with a specific
individual but rather is a property of the behavior that is generalized to a social unit as a
whole, such as a group of knowledge workers in a BPO arrangement (Putnam, 1993).
34


When generalized trust is strong, the effort required for knowledge sharing may not be
salient to knowledge contributors because they believe that knowledge shared is not
likely to be misused by the other party (Davenport and Prusak, 1998). Conversely when
generalized trust is weak, knowledge contributors may find the effort required for
knowledge sharing to be high as they believe that others may inappropriately use their
knowledge. Generalized trust has been viewed as a key factor that provides a context
for cooperation among multiple units (Nahapiet and Ghoshal, 1998). The generalized
trust resource has three components that help define its formation and existence (1) the
establishment and keeping of obligations, (2) giving credit to people for contributions
where credit is due, and (3) using other workers or units knowledge appropriately
(Kankanhalli, Tan and Wei, 2005). Generalized trust is an extremely important IS
resource in the offshore BPO context for effective knowledge exchange because of the
absence of adequate personal knowledge among the collaborators and coworkers in the
client and vendor firms.
Pro Sharing Norms
A norm represents a degree of consensus in the social system (Coleman, 1990).
Norms have the effect of moderating human behavior in accordance with the
expectations of the group or community. Pro sharing norms therefore are likely to
enhance the climate for knowledge sharing and positively impact knowledge
management capabilities in the firm. Components of the pro sharing norms resource
include (1) teamwork collaboration and sharing (Jarvenpaa and Staples, 2000); (2)
willingness and ability to value and respond to diverse views and opinions among staff
and (3) the tolerance for mistakes and failure in the relationships. When Pro sharing
norms are strong, the costs of knowledge sharing may not be a deterrent to knowledge
35


contributors (Kankanhalli, Tan and Wei, 2005). Pro sharing norm enable the workers
from diverse backgrounds and remote firms client and vendor in the BPO to forge
common language, conventions and values and leads to improvements in knowledge
management capabilities in the BPO arrangement. Hence Pro Sharing Norms are an
important component of the social capital resource in the context of offshore BPO.
Infrastructure Resources
Infrastructure resources refer to a firms shared IT assets (e.g., hardware,
software tools, and networks, databases and data centers). They are the foundations for
a firms IT architecture, which is the plan or design that supports multiple service
processes, products and groups of users groups across the firms. From a strategic
resource perspective for building strategic level capabilities, the individual IT asset is
commodity-like, widely available, imitable, and relatively easy to obtain, and is therefore,
incapable of generating long-term economic rents, and generally not found to be a source
of sustained competitive advantage (Wade and Hulland, 2005). Reliable IT infrastructure
resources, however, can ensure the success of IT architecture, which tends to be highly
firm specific and evolves over a long period of time during which gradual enhancements
are made to reflect changing business needs (Karimi, Somers and Bhattacharjee,
2007a).
Knowledge Management Capabilities
Drawing on the research stream in knowledge management capabilities,
knowledge management is defined through the five processing steps involved: (1)
Acquisition, (2) Application, (3) Conversion (4) Transfer and (5) Protection (Szulanski,
1996; Gold, Malhotra and Segars, 2001; Tanriverdi, 2005). The actions that result form
the above consist of one or both parties seeking knowledge and/or providing knowledge
36


in response to the request, such that one or both parties are affected by the experience
(Huang and DeSanctis, 2005, Szulanski, 1996). The facets of knowledge sharing in a
BPO are: (1) one or both parties seeking to acquire knowledge, (2) one or both parties
converting tacit knowledge or pointing to the location of already explicit knowledge in
response to the request, (3) one or both parties transferring the knowledge and (4) the
seeking party applying the new knowledge. This knowledge can be related to either the
outsourced process and/or product/market(s) that are being served by the process.
Finally, knowledge must also (5) be protected against unauthorized access and use on
both sides of the outsourcing arrangement.
BPO Outcomes
Coordination theory offers a vehicle to measure organizational performance
through business processes referred to in the theory as coordination structures
(Malone, 1987). These information flows and coordination structures manifest in three
types of information costs. Hence to measure a process outcome, the measure must
capture the reflection of the information flow of the coordination structure of the process
on the three information costs Production, Coordination and Vulnerability (Ghosh and
Scott, 2006a). There are three types of information costs associated with each
coordination structure:
1. Production costs are the transaction costs of running the outsourced
processes and they measures any efficiencies or deficiencies introduced.
To evaluate BPO outcome, the change in the clients transaction costs after
the BPO implementation must be measured.
2. Coordination costs are the management costs to manage the
communication between the client and the vendor firms, prioritization of
37


activities and the allocation of resources. To evaluate BPO outcome, the
change in the client's coordination costs after the BPO implementation must
be measured.
3. Vulnerability costs are the strategic costs associated with a delayed
response to a changed market situation. The client firm in a BPO is further
removed from the market/end user resulting in missed market knowledge.
This leads to additional vulnerability costs for the client. To evaluate BPO
outcome, the change in the clients vulnerability costs after the BPO
implementation must be measured.
The combination of the three information costs derived from coordination theory
is termed as BPO outcome (BPOC). The survey instruments are listed in Table 4.1.
38


Table 4.1 Survey Instruments
SOCIAL CAPITAL RESOURCE (SCR)_______________________________________________________
SCR1: There is a norm of teamwork among staff in the BPO relationship.__________
SCR2: There is a norm of openness of diverse/conflicting views among staff in
____________the BPO relationship_____________________________________________________
SCR3: There is a norm of tolerance of mistakes among staff in the
____________BPO relationship_________________________________________________________
SCR4: Client and vendor staff in the BPO relationship share common values.
SCR5: Client and vendor staff in the BPO relationship share common goals
SCR6: Client and vendor staff in the BPO relationship have strong group
__________identification_____________________________________________________________
SCR7: Client and Vendor staff in the BPO relationship give credit to people
___________where it is due___________________________________________________________
SCR8: Obligations between organizations are always sustained by staff in the
___________BPO relationship__________________________________________________________
SCR9: Client and Vendor staff in the BPO relationship use others knowledge
___________appropriately_____________________________________________________________
INFRASTRUCTURE RESOURCES (IR)_______________________________________________________
IR1: Appropriate hardware, software, and network infrastructures are in place
___________to support the BPO________________________________________________________
IR2: Appropriate knowledge systems and tools are in place in the infrastructure
___________to support the BPO________________________________________________________
IR3: Access technologies are administered in the infrastructure to allow
___________appropriate access to knowledge ("Item dropped)___________________________
KNOWLEDGE PROCESS CAPABILITIES Acquisition________________________________________
ACQ1: Our organizations have processes for acquiring knowledge from our
______________BPO partners___________________________________________________________
ACQ2: Our organizations have processes for generating new knowledge from
______________existing knowledge in the BPO relationship_____________________________
ACQ3:Our organizations have processes for exchanging knowledge between
_____________individuals in the BPO relationship.____________________________________
KNOWLEDGE MANAGEMENT CAPABILITIES Conversion______________________________________
CNV1: Our organizations have processes for converting knowledge into the
_____________design of future BPO support services___________________________________
CNV2:Our organizations have processes for distributing knowledge throughout
_____________the two organizations in the BPO relationship___________________________
CNV3: Our organizations have processes for integrating different sources and
______________types of knowledge (*Item dropped)_____________________________________
39


Table 4.1 Survey Instruments (cont.)
KNOWLEDGE MANAGEMENT CAPABILITIES Application____________________________________
APL1: Our organizations have processes for applying knowledge learned from
_____________experience____________________________________________________________
APL2: Our organizations have processes to make knowledge accessible to those
_____________who need it___________________________________________________________
APL3: Our organizations have processes to quickly link sources of knowledge
_____________in solving problems___________________________________________________
KNOWLEDGE MANAGEMENT CAPABILITIES Transfer ______________________________________
TRF1:1 am encouraged to discuss my work with employees in other
_____________workgroups____________________________________________________________
_______TRF2: Benefits of sharing knowledge in my organization outweigh the costs
TRF3: Knowledge is freely exchanged freely through email, presentations,
_____________newsletters, etc._____________________________________________________
KNOWLEDGE MANAGEMENT CAPABILITIES Protection_____________________________________
PRT1: Our organizations have processes to protect knowledge from
_____________inappropriate access in the BPO relationship__________________________
PRT2: Our organizations have processes to protect knowledge from
_____________inappropriate use in the BPO relationship_____________________________
PRT3: Our organizations clearly communicate the importance of protecting
______________knowledge____________________________________________________________
BPO OUTCOME Vulnerability________________________________________________________
VUL1: After the BPO, my organization adapts more quickly to unanticipated
______________changes______________________________________________________________
VUL2: After the BPO, my organization can react more quickly to new market
_______________info________________________________________________________________
VUL3: After the BPO, my organizations response to market changes involves
_______________a less lengthy process______________________________________________
BPO OUTCOME Production___________________________________________________________
PRD1: After the BPO, the transaction costs related to these outsourced
_______________processes are lower_________________________________________________
________PRD2: After the BPO, we can serve our customers more effectively now_______
PRD3: After the BPO, less resources are required to operate the outsourced
_______________processes___________________________________________________________
BPO OUTCOME Coordination_________________________________________________________
CRD1: After the BPO, a decreased number of resources are needed to support
______________the outsourced process_______________________________________________
CRD2: After the BPO, a decreased number of managerial contacts are needed
______________to support the outsourced process____________________________________
CRD3: After the BPO, less time is required to get status and updates about
______________the outsourced process_______________________________________________
40


CHAPTER 5
RESEARCH HYPOTHESES
The following subsections build and present the research hypotheses of this
study. A research model is developed to study the impact of social capital and
infrastructure resource on the knowledge management capabilities in the BPO
relationship and the BPO outcome. The research model is illustrated in Figure 5.1.
SCR,
Information \
Systems ''
Resources j
(ISR) /
ACQ f-
ACQ,.
/ Knowledge
/ Management
l Capabilities
>, (KMC)
i, APL j
(
CNV
APL.
CNV,.
\
( TRF ")
BPO
Outcomes
(BPO)

PRT,
VUL
VUL..J
j-----PRD ) pm>i.3
( CRD
CRD,
/ ISR X\
( X ]
\ KMC /
Figure 5.1 Research Model
41


As reported earlier, Wasko and Faraj (2005) found that motivational factors and
organizational enablement impact the knowledge management in the relationship.
Kumar, Dissel and Bielli (1998) described the textile merchants in Prato, Italy. Their
study showed how face-to-face operations and closeness of business relationships have
built enablers for knowledge management in their business practices over the years.
Such close interactions and effective knowledge management can be established in a
non-collocated team as well, when social capital resources are established. Malhotra, et.
al. (2001) reported that the development of common language and norms between non
collocated teams allows better use of collaboration technologies and knowledge
management in a case study of a virtual team at Boeing. Such a virtual environment also
was found to facilitate more knowledge management than a regular collocated group.
In their study of knowledge transfer from consultants to clients, Ko, Kirsch and
King (2005) found the impact of several components on the effective management of
knowledge in inter firm relationships. They identified factors such as communication
encoding and decoding capacity that builds a shared understanding that impacts
knowledge transfer. Szulanski (2000) identifies the importance of "bridging the
communication gap, the coding schemes and cultural conventions" as critical to
overcoming knowledge management stickiness. This study also claims that knowledge
transfer occurs as social capital is built up and documented. Other studies have
investigated virtual communities or remote networks of practice (Brown and Duguid,
2001) and found that the social capital resource in the relationship significantly impacted
the quality of knowledge contributions. Social capital builds trust and commitment and is
a facilitator for collective action when multiple parties are engaged.
42


Knowledge management is more likely when social relationships and generalized
trust are strong (Szulanski, 1996; Ko, Kirsch and King, 2005). Social capital resources
allow strategic networks to be built that facilitate knowledge management (Gulati, Nohria
and Zaheer, 2000). These networks are seen to encourage group identification. Inkpen
and Tsang (2005) also found that behavioral factors play a significant role in knowledge
transfer in these social networks. The stream of research in partnerships and inter-
organizational learning has also identified that social capital resources in the relationship
play a role in knowledge transfer in both formal and informal settings (Malhotra, Gossain
and El Sawy, 2005; Argote, 1999). Such collective action encourages the development
and management of knowledge as group identification is built up.
The above arguments support the first hypothesis (H,), which is stated as
follows. Note that H, is shown in Figure 5.1 as a link from Social Capital Resources
(SCR) to Information Systems Resources (ISR), however, the hypothesized relationship
that is tested is a link from SCR to Knowledge Management Capabilities (KMC). Please
refer to the direct effects model in Figure 8.1.
Hi- Social capital resources have a positive association with knowledge
management capabilities in the BPO organization.
IT infrastructure resources have been linked to a firms ability and willingness to
develop innovative business applications and knowledge management capabilities
(Davenport and Prusak, 1998). Such applications can be in quite far reaching areas such
as business intelligence for knowledge discovery, case based reasoning for knowledge
application, content management and servers for knowledge codification, etc. They
support and fully facilitate the firms knowledge sharing capabilities (Karimi, Somers and
Bhattacharjee, 2007a). IT infrastructure resources are critical in developing, building, and
43


assimilating IT capabilities, and in enhancing ITs productive value by aiding
implementation, simplifying system integration across diverse applications, and creating
economies of scale and scope in system maintenance. In contrast, lack of IT
infrastructure resources severely restricts a firms IT capabilities and increases the costs
of building or supporting IT (Ray, Muhanna and Barney, 2005). Clearly adequate
infrastructure resources (e.g., hardware, software, network, and server and database
technologies) need to be planned, put in place, and reliably available well in advance
both for pre-and-post BPO implementation stages.
The above arguments support the second hypothesis (H2), which is stated in the
alternate form as below. Note that H2 is shown in Figure 5.1 as a link from Infrastructure
Resources (IR) to Information Systems Resources (ISR), however, the hypothesized
relationship that is tested is a link from IR to Knowledge Management Capabilities (KMC).
Please refer to the direct effects model in Figure 8.1.
H2- Infrastructure resources have a positive association with the knowledge
management capabilities in the BPO organization.
Resources are essential raw materials for capability-building, and their availability
determines a firm's ability to build such capabilities, which are often critical drivers of firm
performance. Resource synergies and the significance of integration, combination, and
co-specialization of resources have been emphasized in past research (Tanriverdi and
Venkataraman, 2005). Further, some capabilities may derive from a contribution of a
single resource, while others may require highly complex interactions involving the
cooperation of many different resources (Ray, Muhanna and Barney, 2005).
A review of the prior research on BPO shows support for the combined IS
resources to impact the knowledge management capabilities. Levina and Ross (2003)
44


have used economic theory to suggest that outsourcing relationships are successful in
the cases where the client firm maintains proper organizational infrastructures to manage
the relationship between the client and vendor and develop complementarities. This
study has also reported that the information processing requirements for the client
organization need to be matched with the level of uncertainty facing the organization in
the BPO. Lacity and Willcocks (1998) studied the allocation of resources and found that
infrastructure resources must be properly distributed to meet organizational requirements
and emphasize contributions to knowledge repositories. Outsourcing relationships are
usually set up to have a loose relationship between the firms. Such a loose alliance
creates a fragmented internal environment, which can hamper the knowledge flow and
cause the BPO client to become vulnerable to missing key knowledge from the vendor
side similar to loosely coupled environments (Orton and Weick, 1990), which require IS
resources to support the knowledge management system in the BPO.
When a BPO is pursued, additional infrastructure needs to be established so that
the organization can effectively exchange knowledge within the BPO context (Levina and
Ross, 2003). Hence the effectiveness of the knowledge management capabilities of the
firm after the BPO deployment will depend on the effective pairing of other relationship
based resources with IT infrastructure resources that is accessible to both sides of the
BPO client and vendor. Similarly, because of resource synergies, individual dimensions
of IS resources should not be viewed in isolation. As a collective, mutually reinforcing,
and a higher order factor structure, which accounts for the relationships among social
capital resources and IT infrastructure resources, IS resources are expected to have a
combined impact on building knowledge management capabilities. The above arguments
support the third hypothesis (H3), which is stated as:
45


H3- Information Systems (IS) Resources have a positive association with
knowledge management capabilities in the BPO organization.
BPO vendors allocate their resources over multiple clients. Thus, BPO
environments are unique in that client initiated knowledge management capabilities will
not only benefit the initiating client and its vendor, but other clients of that vendor as well.
However, knowledge created through the mobilization of social capital resource is
retained in the work group (Nahapiet and Ghoshal, 1998). Such assets are jointly
owned and symmetrically transferred within the BPO work team established by the client
and vendor. Hence, positive benefits can be measured in the client organization (Ghosh
and Scott, 2006a). It is therefore advisable to measure the impact of knowledge
management capabilities on the process cost structure of the client.
Knowledge management capabilities reduce the efforts involved with
coordination of BPO activities (Ghosh and Scott, 2006a). Productions costs related to
the operation of the process that is outsourced are also lowered with the increase in
knowledge transfer since improved performance is to be expected from the vendor as
they become more knowledgeable. These benefits arise as the vendor can access
knowledge and reuse it more rapidly (Watson and Hewett, 2006), hence lower production
costs result.
Creating opportunities for knowledge management results in more informal
exchanges of information among personnel in the two firms. This effectively improves
coordination capabilities (Ghoshal, Korine and Szulanski, 1994). The research stream in
project management also identifies the flow of knowledge as a critical resource in
coordination activities (Snider and Nissen, 2003). The greater the scope and frequency
of knowledge management, the smoother the project coordination tasks such as resource
46


allocation and task prioritization. In a BPO, coordination costs will be lowered as
knowledge management increases.
Outsourcing arrangements are usually set up to have a loose relationship
between the firms. Such a loose alliance creates a fragmented internal environment,
which can hamper the partnership culture and information flow and cause the BPO client
to be vulnerable to market changes (Lacity and Willcocks, 1998). In a typical BPO the
client often relinquishes some of the customer interactions that are now managed by the
vendor. Mechanisms need to be put in place to facilitate the management of customer
knowledge that is collected by the vendor in their customer interactions back to the client.
Therefore, with an increase in knowledge management, it is anticipated that vulnerability
costs will be also be lowered (Ghosh and Scott, 2006a).
Therefore the impact of increasing the knowledge management in the BPO
relationship is to reduce all three information costs which improves the BPO outcome.
The above arguments support the fourth hypothesis (H4), which is stated as:
H4- Knowledge management capabilities have a positive association with
BPO outcomes in the BPO organization.
Capabilities are special types of resources that are non transferable between
firms and whose purpose is to improve the productivity of other resources. Resources
and capabilities complement each other and firms need to effectively select and match
the right resources to build the needed capabilities. Moreover to support building those
capabilities, the organization needs to effectively deploy those resources as well
(Melville, Kraemer and Gurbaxani, 2004). In addition, certain organizational capabilities
may be complementary to the IS resources in generating IT business value. Melville,
Kraemer and Gurbaxani (2004) examined the nature of resource synergies and further
47


suggested that human IT resources are complementary to IT infrastructure resources and
they both create economic value for a foe a firm by conferring operational efficiencies that
vary in magnitude and type depending upon the organizational and technological context.
Their study further suggested that (1) there are complementary relationships between
certain IS resources in generation of IT business value, and (2) the existence and
magnitude of the complementarity between any two specific instantiations of these
resources varies depending upon the organizational and technological contexts.
IT infrastructure and e-commerce capability, for example, have been shown to
have positive complementarity effects on some measures of firm performance (Zhu,
2004). These studies suggest that IT infrastructure resources can provide the platform to
deploy knowledge management capabilities in the BPO. Further the effectiveness of
those knowledge management capabilities as seen on the BPO outcome may be
enhanced by the amount and the combination of IS resources of social capital and
infrastructure that are deployed.
The social capital resource affects the BPO outcome by impacting the
information costs of production, coordination and vulnerability through the knowledge
management capabilities. Social capital can reduce the cost of an item of work in the
outsourced process by economizing on information and coordination effort (Nahapiet and
Ghoshal, 1998). A study by Koh, Ang and Straub (2004) has stressed the need to have
psychological contracts, i.e., social capital within individuals that can drive the behavior of
vendor personnel, such as such as sharing knowledge among individuals in the BPO
arrangements.
The same arguments have been found for the Infrastructure resource, which
enable knowledge management capabilities by allowing the storage, search,
48


administration of access rights and retrieval of knowledge resources both within and
across organizations (Scott, 2000). Infrastructure supports building inter organizational
capabilities, however the strength of the capability and its impact on the performance of
the inter-firm BPO arrangement depends on the relationship resources deployed (Scott,
1998). A simple IT system such as a transaction system is enough to capture information
but is inadequate for knowledge management, which needs human resources in
combination with those infrastructure resources to build, assimilate and leverage those
knowledge entries into the business processes (Scott, 1998). Hislop (2002) points out
that technology infrastructure resources can facilitate capabilities for explicit knowledge,
while relationship resources are needed for tacit knowledge capabilities. Clearly a
complementary relationship exists among the IS resources social capital resources and
infrastructure resources and KM capabilities on the BPO outcome. Any weakness in
these IS resources will undermine the value derived from building those knowledge
management capabilities in the BPO.
These arguments suggest that there is a strong theoretical foundation to support
a complementary relationship among IS resources and knowledge management
capabilities that impacts BPO outcome. Any weakness in IS resources social capital
and infrastructure can reduce the benefits produced by knowledge management
capabilities on BPO outcomes. Likewise higher IS resources will promote more effective
knowledge management and will lead to better BPO outcomes. Therefore, the fifth
hypothesis (H5) can be stated as:
H5- Greater IS Resources in conjunction with higher knowledge
management capabilities are associated with improved BPO outcomes in
the BPO organization.
49


CHAPTER 6
RESEARCH METHODOLOGY
Research Design
The methodology used is a field study. The study was conducted in two stages.
The first stage involved a pilot study to evaluate the survey the wording and descriptive
explanations on the survey.
The second stage involved a survey to collect data from client side employees.
Knowledge workers in multiple supervisory groups involved in the operation of customer
service and network consulting processes within the client organization in a high
technology outsourcing environment were surveyed. The survey measured the actual
presence of IS resources and the knowledge management capabilities that take place
between the two organizations in the BPO arrangement. The survey also measured the
outcomes of outsourcing reflected in the three types of costs: vulnerability, coordination
and production costs. The quantitative data that was collected from the surveys was
analyzed with PLS-Graph software tool using the Partial Least Squares method of
structural equation modeling (Chin, 1998).
Study Variables
The independent variables for this study are the Social Capital Resources (SCR)
and the Infrastructure Resources (IR), which form a formative construct termed IS
Resources (ISR). ISR is a second order formative construct, which is formed by the
above two first order constructs social capital resources and infrastructure resources.
The dependent variable is the BPO outcomes (BPO) as measured by the three
coordination costs production, coordination and vulnerability. Knowledge management
capabilities (KMC) is the additional variable measured in the research model. The data
50


was collected at the worker level in the client organization. Knowledge workers rated
their perceptions of the variables. Items were scored on a 5 point Liekert scale of
Strongly Agree (5), Agree (4), Neutral (3), Disagree (2) and Strongly Disagree (1).
Measurement of Research Variables
All research instruments used in the study were adopted from prior research
studies. Hence they were pre-validated research instruments that only needed content
validation for their use in the BPO domain. The Knowledge management capabilities
(KMC) construct is a second order factor that is measured with five reflective first order
constructs with 3 indicators each. The five constructs are: knowledge acquisition (ACQ
3), knowledge conversion (CNV^), knowledge application (APL^), knowledge protection
(PRT,^) and knowledge transfer (TRF^). The individual items were adopted from the
knowledge management capabilities model developed and validated by Tanriverdi (2005)
and Gold, Malhotra and Segars (2001). The components of knowledge management
capabilities in a BPO are: (1) one or both parties seeking to acquire knowledge, (2) one
or both parties converting tacit knowledge or pointing to the location of already explicit
knowledge in response to the request, (3) one or both parties transferring the knowledge
and (4) the seeking party applying the new knowledge. This knowledge can be related to
either the outsourced process and/or product/market that is being served by the process.
Finally, knowledge must also (5) be protected against unauthorized access and use on
both sides of the outsourcing arrangement.
IS resources (ISR) is a formative index formed by two reflective first order
constructs social capital resources (SCR) and IT infrastructure resources (IR). Social
capital resources (SCR) were measured by nine reflective items, three each from each of
the three dimensions pro sharing norms, group identification and generalized trust. The
51


definition of the social capital construct by Nahapiet and Ghoshal (1998) specifically
states that the these dimensions cannot be easily separated and the that the social
capital manifests into these three dimensions Three items reflect the Pro sharing norms
dimension (SCR^s) which tapped into the workers acceptance and use of cognitive
norms in knowledge sharing in the BPO. The group identification dimension was reflected
using three items (SCR4.6) that tapped into the strength of group unity and common goals
among the staff on the BPO. The generalized trust dimension of SCR was reflected
using three items (SCR7.9) that probed into the relational aspects, which refers to the
history of interactions among the people and how it influences their trust behavior. These
instruments were adopted from the social capital study on electronic knowledge
repository usage developed and validated by Kankanhalli, Tan and Wei (2005). IT
infrastructure resources (IR) were measured using three items (IR1.3) that assessed
whether hardware, software, network, and the necessary server and database
technologies, knowledge tools and access restrictions were in position prior to the BPO
deployment.
52


CHAPTER 7
DATA COLLECTION
Collection Methodology
The client organization chosen is a multinational technical support organization.
The organization is a leader in the support of multi-vendor networking equipment, with
network management and service capabilities. Network service offers include design,
installation, monitoring and break-fix support. The product introductions, field trials,
network design and installation services were outsourced to an offshore vendor to
increase available headcount for these processes, reduce operating costs through labor
arbitrage and provide in-region internationally located technical personnel.
Organizational Practices for Knowledge Management
The chosen organization was found to have several KM systems and practices to
provide bi-directional transfer of knowledge from the client to the vendor and vice versa.
In this BPO relationship, the client personnel are more experienced in the work domain
than the vendor personnel. The client organization realized that to make the BPO
effective vendor personnel needed to be mentored and trained in the domain. Hence, the
heavier knowledge transfer involved the training and development of vendor personnel to
prepare them for the technical work involved in the BPO. The lesser knowledge transfer
occurred from the vendor to the client in capturing customer site-specific information, as
the vendor is closer to the installation locations.
The mentoring resources on the client side were limited and needed to be
managed effectively. To serve the two fold goals of providing training to the vendor
personnel as well as supporting the bi-directional knowledge transfer, the client decided
to institute practices to build social capital. A selection program was instituted, whereby
53


vendors could apply for client mentoring and assistance on an upcoming project. The
client established a review board to screen each request for knowledge potential and
optimal fit for the goals. If a project is selected, then personnel from the vendor and client
work collaboratively on the project over a span of 4-6 weeks.
Advantages are seen in building one or more social capital resources in terms
of exchange of norms, work practices, common language and standards as well as tacit
knowledge that is difficult to codify and contextualize. Relational resources described
earlier are built up as obligations are set and met, resulting in the development of trust,
understanding of cultural diversity and establishment of common identification and joint
ownership for work. The latter was critical, as the BPO goals were broader than cost
savings and also involved building competencies. Project meetings are held frequently
and the client personnel served as mentors in the relationship thus there was an
expectation of value to be obtained from the process on the part of the participating
vendor staff. Likewise, the client participant was able to build up site specific and
international expertise.
Both parties, client and vendor, are motivated to participate in the program to
improve network service quality and minimize down stream customer network issues.
54


Outsourced Processes
Survey respondents fell into one of two services processes customer service or
networking consulting. Both processes shared the IT infrastructure, knowledge
management capabilities and worked with the same vendor firm on the offshore side of
the BPO relationship. The flow of these two processes are given below.
Customer Service Process
Customer Service consists of fielding customer reported troubles, collecting
information and investigating and resolving those customer issues. This process
currently has domestic staff as well as staff at the vendor site offshore. The client staff
has the added responsibility of training and mentoring the vendor staff so they come up
to speed on the process and work. The primary reasons for pursuing outsourcing for this
process include: (1) to provide local staff for supporting regional customers in their time
zone, (2) creating a 24 hour support day by using global resources, (3) reducing the
operating expenses of running the process by utilizing cheaper offshore staff, (4)
capturing and institutionalizing localized knowledge from vendor staff and (5) increasing
the headcount of trained personnel on the process.
The steps of the process are:
1. Receive customer problem notification either in the form of a customer call, an
email, web based problem report.
2. The knowledge worker then makes contact with the customer to collect additional
information about the problem, the products involvement, the customers
environment, etc.
3. The worker then examines the problem, including various research activities
such as looking through product manuals, setting up and working on lab
55


resources, consulting various knowledge bases, listservers, FAQ, etc and
discussing with experts.
4. The worker can repeat steps 2 and 3 multiple times as more information is
needed or as details are collected. This is very much adhoc processing, where
very detailed and sensitive information about the customer is collected. The
steps to be followed can be very different based on the information collected.
5. The worker has to take extreme precaution to not disrupt the customer's network
too much and as well as not to hamper working sections of the network or
products that are in service. The information is quite sensitive as they represent
configuration data from working product installations.
6. Once a solution has been identified, the worker then performs the necessary
steps to solve the customer's problem.
Network Consulting Process
Network Consulting consists of fielding customer requests for information and
delivery of proposals and sales configurations in support of the customer request. The
process involves collecting information at a high level and understanding the customers
needs to develop a proposal to be presented to the customer. This process currently has
domestic staff as well as staff at the vendor site offshore. The client staff has the added
responsibility of training and mentoring the vendor staff so they come up to speed on the
process and work. The primary reasons for pursuing outsourcing for this process
include: (1) to provide local staff for supporting regional customers in their time zone, (2)
reducing the operating expenses of running the process by utilizing cheaper offshore
staff, (3) capturing and institutionalizing localized knowledge from vendor staff and (4)
increasing the headcount of trained personnel on the process.
56


The steps of the process are:
1. Receive customer request for information and product/solution suitability either
in the form of a customer call, an email, web based request.
2. The knowledge worker then makes contact with the customer to collect additional
information about the request, their environment, etc.
3. The worker then examines the request using various configuration and design
templates, product manuals, consulting various knowledge bases and discussing
with experts.
4. The information collected by the worker is generally high level and often a best
guess" or approximate and hence is not as sensitive to the customer that would
require extra security measures.
5. The worker generates a well structured proposal in a standard format to present
the information and solution options to the customer.
Survey Results
The final survey consisted of 4 demographic questions and 36 items. The survey
was performed using an online survey. Email solicitations were sent to 200 client
knowledge workers with a link to the survey, which was active for one week. A
convenience sample was drawn from a list of professional contacts. The 200 employees
fell into 14 supervisory groups. Each group was responsible for supporting one or more
products or product families or market segment. A total of 119 employees responded
with completed surveys for a response rate of 60%. The final set of 119 respondents
were not identified and therefore cannot be traced to a supervisory group. The survey
requested that they identify which one of the two processes that they followed in their
work customer service or network consulting.
57


The respondents fell into one of two processes customer service (72
respondents) and network consulting (47 respondents). Comparisons of demographic
factors of years of experience, years on the current job, years of education and gender
was made between the respondents on the two processes and no significant difference
were found using T-Tests (Table 7.1).
The descriptive statistics for the data is shown in Table 7.2. The mean statistic,
standard error and range and the skewness and kurtosis measurements suggest that the
measurements do approximate normal distributions.
58


Table 7.1
Demographics across Processes
Characteristic Custorm Service it [72) Network Consulting (47) T-Value
Mean Std Mean Std
Years on Job 5.40 2.84 5.51 3.66 .18 (0.857)
Professional Experience 9.01 6.69 8.35 6.95 -.52 (0.604)
Years of Schooling 15.17 2.55 15.09 2.84 -.16(0.871)
Gender LL 00 2 24) M (35), F (12) X*= 1.18; p=0.277
59


Descriptive Statistics
N Ranae Minimum Maximum Mean Std. Variance Skewness Kurtosis
Statistic Statistic Statistic Statistic Statistic Std. Error Statistic Statistic Statistic Std. Error Statistic Std. Error
PSN1 119 4 1 5 2.00 .097 1.054 1.112 .414 .222 -.708 .440
PSN2 119 4 1 5 3.24 .093 1.014 1.029 .105 .222 -.026 .440
PSN3 119 4 1 5 2.82 .078 .850 .723 .432 .222 -.438 .440
ID1 119 4 1 5 2.99 .097 1.054 1.110 .459 .222 -.900 .440
ID2 119 4 1 5 3.29 .081 .886 .706 -.320 .222 -.819 .440
ID3 119 4 1 5 2.68 .086 .938 .880 .557 .222 -.599 .440
GT1 119 4 1 5 3.22 .072 .783 .613 .838 .222 .953 .440
GT2 119 4 1 5 3.29 .083 .903 .816 -.389 .222 -.023 .440
GT3 119 4 1 5 3.35 .076 .829 .688 -.743 .222 -.687 .440
IR1 119 4 1 5 3.38 .107 1.164 1.356 -.617 .222 -.613 .440
IR2 119 4 1 5 3.34 .100 1.091 1.191 -.505 .222 -.540 .440
IR3 119 3 2 5 3.86 .072 .784 .615 -.278 .222 -.315 .440
ACQ1 119 4 1 5 3.08 .090 .984 .969 -.425 .222 -.862 .440
ACQ2 119 4 1 5 2.79 .086 .938 .079 .057 .222 -.149 .440
ACQ3 119 4 1 5 3.47 .077 .842 .709 -1.162 .222 .552 .440
CNV1 119 4 1 5 3.32 .082 .892 .795 -.386 .222 -.824 .440
CNV2 119 4 1 5 3.39 .095 1.035 1.071 -.250 .222 -.542 .440
CNV3 119 4 1 5 2.94 .094 1.026 1.056 .310 .222 -.873 .440
APL1 119 4 1 5 3.02 .070 .854 .729 .051 .222 -1.035 .440
APL2 119 4 1 5 3.16 .070 .766 .507 -.003 .222 -.208 .440
APL3 119 4 1 5 2.94 .085 .932 .869 .055 .222 -.883 .440
PRT1 119 4 1 5 3.60 .086 .942 .007 -.594 .222 -.109 .440
PRT2 119 4 1 5 3.22 .077 .835 .698 .101 .222 -.253 .440
PRT3 119 4 1 5 3.55 .085 .927 .859 -.602 .222 -.090 .440
TRF1 119 4 1 5 3.10 .087 .945 .892 .006 .222 -.740 .440
TRF2 119 4 1 5 3.25 .096 1.051 1.105 .100 .222 -.785 .440
TRF3 119 4 1 5 3.01 .097 1.054 1.110 .248 .222 -.576 .440
VUL1 119 4 1 5 2.85 .108 1.183 1.401 .267 .222 -.913 .440
VUL2 119 4 1 5 2.63 .096 1.049 1.099 .343 .222 -.708 .440
VUL3 119 4 1 5 2.76 .101 1.102 1.215 .211 .222 -.899 .440
PRD1 119 4 1 5 3.34 .065 .925 .855 -.416 .222 -.613 .440
PRD2 119 4 1 5 2.99 .103 1.124 1.263 -.020 .222 -.836 .440
PRD3 119 4 1 5 2.90 .103 1.123 1.261 .340 .222 .812 .440
CRD1 119 3 2 5 3.50 .070 .852 .727 -.514 .222 -.578 .440
CRD2 119 4 1 5 3.33 .092 1.001 1.002 -.390 .222 -.368 .440
CRD3 Valid N (listwise) 119 119 4 1 5 3.59 .087 .951 .905 -.768 .222 .148 .440
Table 7.2 Descriptive Statistics of Items


Knowledge Resources
Data was collected on the usage of the following knowledge management
systems over an 8 month period, while the knowledge management system was
operational.
(1) Listservers, Discussion Boards, FAQ that captured threads of discussions
on topics raised by team members and their subsequent contributions. On average 29
unique issues or threads of discussion have been activated per month over the period.
(2) Checklists to guide technical support work from past experience and ensure
that adequate data collection and situational analysis is being done. A total of 33
checklists have evolved in the same period ranging from building best practices in
directed tasks such as configuring phones to open-ended tasks such as collecting
customer information for network designs.
(3) Lessons Learnt Lists to ensure that new rules of thumb and experience is
captured and shared for future use. On average 54 lessons learnt bullet points were
documented per month over the eight-month period.
(4) Training Presentations Periodically, as project experience is built up,
training materials are developed by scouring the listservers, boards, FAQ. Monthly
training presentations created by the client personnel are imparted to all vendor
personnel. Presentations include all the listen/ threads and their resolutions, the list of
lessons learnt and pointers to any checklists or process/product document that is
considered a must read.
61


CHAPTER 8
DATA ANALYSIS AND RESULTS
Partial Least Squares, PLS-Graph, v3.0, Build 1126 (Chin, 1995) was used to
test the research model. PLS supports the hierarchical component approach for
modeling second-order factors in which the second-order factor is measured using the
first-order factor scores (from the measurement model) as manifest indicators of the
second order construct. In PLS, such factors can either be formative or reflective. PLS
was particularly well suited for analysis of the data given its flexibility to handle second-
order constructs and constructs with both reflective and formative indicators. In the
research model, all the first order factors (SCR, IR, ACQ, APL, CNV, TRF, PRT, VUL,
PRD and CRD) are reflective. Due to limitations of the analysis tool, the SCR was
measured as a single first order factor of nine items, rather than a reflective second order
factor of 3 first order items representing the three dimensions of SCR pro sharing
norms, generalized trust and group identification.
Reflective indicators are believed to reflect the unobserved, underlying
construct they posit to represent, with the construct giving rise to (or causing) the
observed measures. In contrast, formative indicators have several characteristics that
cogently distinguish them from reflective indicators. First, formative indicators form the
construct as a composite (Jarvis, Mackenzie and Podsakoff, 2003). With formative
indicators omitting an item is omitting a part of the construct, whereas reflective indicators
are interchangeable, and the removal of an item does not change the nature of the
underlying construct. Finally, reflective indicators of the same construct should show a
high correlation with one another to ascertain construct validity. Formative indicators are
not assumed to reflect the same underlying construct, that is, they can be independent of
62


one another and measuring different factors. In fact, formative indicators of the same
construct can have positive, negative, or no correlation with one another.
Two models were used in the analysis a direct effects model (Figure 8.1) where
SCR and IR interacted directly with KMC and a second order model (Figure 8.2), where
an ISR construct was introduced to assess the combination and interaction of the two IS
resources SCR and IR used in this study.
nt,.
SCR

/ Social Capital
: Resources ,
V (SCR) /
^ ""''v
Infrastructure
1 Resources ,
\ (IR) /
VUL V-
Dru \ -----.N
Outcomes J----prd
' (BPO) J '---
( CRD }-
____. / Knowledge
j( ACQ )-------( Management \
S .... > \ Capabilities )
acqi-3 ^, ckmc) y
X
/ ( CNV ^ "
apl,.3 tV TRF
( PRT )
- VUL,.3
)* PRDl-3
crd,.3
Figure 8.1 Direct Effects Model
63


Figure 8.2 Second Order Model
64


The PLS modeling approach involves two steps: first, validating the
measurement model and then fitting the structural model. The former is accomplished
primarily by reliability and validity tests of the measurement model, followed by a test of
the explanatory power of the overall model by assessing its explained variance, and the
testing of the individual hypotheses (structural model). A bootstrap re-sampling
procedure was conducted using 200 samples and path coefficients were re-estimated
using each of these samples (Chin, 1995). In PLS, validation is done using the
Composite Reliabilities (CR) and Average Variance Extracted (AVE) from the
measurement model in PLS-Graph (Chin, 1998).
To assess reliability and validity the block of indicator's composite reliabilities and
the average variance explained (AVE) are calculated by PLS for each construct. The
composite reliabilities should be greater than 0.7. The AVE measures the variance
captured by the indicators relative to measurement error and it should be greater than 0.5
to justify using a construct. Moreover, the square root of each constructs AVE must be
greater than the correlation of the construct to the other latent variables to ascertain
discriminant reliability.
The results to justify using the construct are shown in the Table 8.1 (Direct
Effects model) and Table 8.2 (Second Order model) and indicate adequate composite
reliabilities (CR) and AVEs. Tables 8.1 and 8.2 show the discriminant reliability of each
construct. All values follow the above rule and hence discriminant reliability is exhibited
by the constructs.
65


Construct C.R. AVE Correlation of Constructs and Square root of AVE
SCR IR ACQ APL CNV PRT TRF VUL PRD CRD
SCR .883 .478 .691
IR .873 .782 .090 .889
ACQ .898 .745 .458 .153 .863
APL .800 .581 .518 .067 .302 .762
CNV .846 .733 .526 .085 .357 .357 .856
PRT .859 .671 .425 .124 .247 .101 .514 .819
TRF .923 .800 .679 .009 .472 .569 .487 .352 .894
VUL .860 .685 .625 .102 .357 .645 .395 .367 .512 .828
PRD .871 .693 .590 .083 .361 .545 .374 .342 .411 .465 .832
CRD .929 .815 .037 .005 .316 .056 .220 .108 .086 .110 .316 .903
SECOND ORDER Constructs -Composite Reliability, AVE and Square Root of AVE
Construct C.R. AVE Square Root of AVE
Knowledge Management Capability (KPC) .882 .365 .604
BPO Outcome (BPOC) .837 .381 .617

O
Q.
CD
Table 8.1 Correlations, CR and AVE for Direct Effects


05
-si
Construct C.R. AVE Correlation of Constructs and Square root of AVE
SCR IR ACQ APL CNV PRT TRF VUL PRD CRD
SCR .889 .484 .695
IR .875 .782 .084 .884
ACQ .898 .745 .401 .123 .863
APL .800 .581 .606 .067 .302 .762
CNV .846 .733 .420 .110 .568 .357 .856
PRT .859 .671 .410 .123 .247 .101 .513 .819
TRF .923 .800 .655 .009 .472 .569 .487 .352 .894
VUL .860 .685 .652 .102 .357 .644 .395 .367 .512 .827
PRD .871 .693 .560 .083 .362 .544 .374 .342 .411 .465 .832
CRD .929 .814 .004 .005 .187 .056 .220 .107 .086 .110 .315 .902
SECOND ORDER Constructs -Composite Reliability, AVE and Square Root of AVE
Construct C.R. AVE Square Root of AVE
Information Systems Resources (ISR) N/A N/A N/A
Knowledge Management Capability (KPC) .882 .365 .604
BPO Outcome (BPOC) .837 .381 .617
o
Q.
CD
Table 8.2 Correlations, CR and AVE for Second Order


Estimation of Internal Consistency
The survey employed multi-item scales to measure the first-order factors. The
measurement properties for the reflective constructs were examined by conducting
confirmatory factor analyses using PLS. To assess the internal consistency of the
reflective first-order factors of SCR, KMC and BPO, coefficient alpha and composite
reliability measures were used. Accordingly, as seen in Table 8.3, coefficient alpha
values ranged from 0.633 to 0.883. Likewise, the composite reliabilities for all measures
were high ranging from 0.800 to 0.923. Compared with coefficient alpha, which provides
a lower bound estimate of internal consistency, the composite reliability is a more
rigorous estimate of the reliability (Chin, 1995). The recommended levels for establishing
a tolerable reliability are above the 0.70 threshold and above 0.80 for strong reliability.
Consequently, evidence for internal consistency and the scales reliability are supported
by these results.
68


Table 8.3
Internal Consistency Measures
Measure # Items Coefficient Alpha Composite Reliability
IS Resources (ISR)
Social Capital Resources (SCR) 9 0.856 0.883
Infrastructure Resources (IR) 2 0.849 (IR3 dropped) 0.875
Knowledge Management Capability (KMC)
Acquisition (ACQ) 3 0.831 0.898
Protection (PRT) 3 0.764 0.859
Application (APL) 3 0.650 0.800
Transfer (TRF) 3 0.875 0.923
Conversion (CNV) 2 0.633 (CNV3 dropped) 0.846
BPO Outcomes (BPO)
Vulnerability Costs (VUL) 3 0.763 0.860
Production Costs (PRD) 3 0.779 0.871
Coordination Costs (CRD) 3 0.883 0.929
69


Dimensionality, Convergent, and Discriminant Validity
In the direct effects model (Figure 8.1), the SCR and IR constructs are
conceptualized to have direct relationships with KMC by testing the direct effects of each
of the first-order factors on KMC, whereas the second order model suggests an explicit
multidimensional structure (ISR) whereby SCR, and IR combine to form an emergent
force (ISR) that in turn can affect KMC.
As indicated by the results in Table 8.4, all the loadings were statistically
significant based on t-statistics generated from running a bootstrap on the data, except
for one (BPO -> CRD). To model the complex character of a firms ISR, which was
operationalized effectively in a formative way by a composite across different, unique
sources of the construct SCR and IR. As a result, a second order model was needed
to understand and define how firms use ISR as a combination of SCR and IR.
70


Table 8.4
Structural Model Results
Path Direct Effects Model Second Order Model
Structural Model Hypothesized Relati onships
SCR - KMC (Hi) 0.809 (13.479) * N/A
IR KMC(H*) 0.037 (0.285) N/A
ISR KMC(H3) N/A 0.732 (7.952) *
KMC -> BPO (H4) 0.640 (5.045) * 0.654 (3.670) *
ISR x KMC -> BPO (Hs) N/A 0.0232 (4.579) *
Measurement Model
ISR<- SCR N/A 0.989 (9.814)*
ISR KMC ->ACQ 0.725 (7.049) * 0.724 (6.795) *
KMC -> CNV 0.776(10.243)* 0.775(10.502)*
KMC -> APL 0.663 (6.670) * 0.664 (6.599) *
KMC -> PRT 0.564(1.812)**** 0.566(1.899)****
KMC -> TRF 0.850 (13.045) * 0.852 (16.975) *
BPO -> VUL 0.762(4.605) * 0.752 (4.244) *
BPO -> PRD 0.833 (12.427) * 0.851 (11.181) *
BPO - CRD 0.451 (1.531) 0.571 (1.595)
Dependent Variables R2
KMC 0.6606 0.5360
BPO 0.4272 0.4275
BPO w/ Moderator Term N/A 0.4507
* p<0.001; ** p<0.01; *** p<0.05; ****p<0.10
Notes: Parameter estimates are standardized with t-values
71


Second-Order Model
As shown in Figure 8.2, the second order model specification hypothesizes ISR
as a second-order formative construct formed by two first-order factors made up of SCR
and IR. Formative constructs are conceived to be caused by the underlying
measurement items where each lower-order item represents a distinct contribution to the
higher-order latent construct (Jarvis, Mackenzie and Podsakoff, 2003). Although it was
possible that the factors might be highly interrelated enough to be treated as reflective
factors, this was not supported by the results. In a formative model, SCR and IR may be,
but need not, be highly correlated, and these results suggest this they are not highly
correlated (Table 8.5).
Table 8.5 Correlation Matrix Factors in ISR Second Order Model
SCR IR
SCR Pearson Correlation 1 .000
Sig. (2-tailed) 1.000
N 119 119
IR Pearson Correlation .000 1
Sig. (2-tailed) 1.000
N 119 119
A reflective model would have very high correlations among the first-order factors
(greater than 0.70), which was not the case. Thus a formative model seems more likely
for ISR. Logically, a change in SCR, for example, does not necessarily imply an equal
change in IR, hence a reflective model is less likely. Moreover, the effective use of the
resources is likely to change over time and be affected in a different way by other factors.
72


As such, the sub-constructs of ISR form a higher-order formative model that most
accurately and parsimoniously captures the multi-dimensional nature of ISR.
Before testing to determine if the second-order formative model of ISR (Figure
8.1) is a better fit to the data than the first-order factor model (Figure 8.2), models
statistics for reliability and convergent and discriminant validity was used to verify the
measurement models. The results for the two models are shown in Tables 8.1 and 8.2.
These metrics were similar and provided support for reliability and convergent and
discriminant validity for all constructs.
Hypothesis Testing
The adequacy of the psychometric properties in the measurement model allowed
further testing the direct effect hypotheses (H, and H2) using the direct effects model
depicted in Figure 8.1.
As shown in Table 8.4 all first-order factors of KMC were significantly related with
knowledge management capabilities, KMC as well as all first order factors of BPO were
significantly linked with the higher order BPO construct, except CRD. While no minimum
threshold value for weights has been established, the statistical significance of the
weights can be used to determine the relative importance of the indicators in forming the
construct. As the interpretation of the weights is similar to the beta coefficients in a
standard regression model, it is usual to have lower absolute weights as compared to
loadings. For hypothesis, H^the results show that the effect of social capital resources,
SCR ((3 = 0.809; t = 13.479; p < 0.001) was a lot more important than the effect of
Infrastructure resources, IR (P = 0.037; t = 0.285; p < 0.05) on the knowledge
management capabilities construct, KMC. Hypothesis H! is therefore supported by the
data and SCR has a significant influence on KMC. However, hypothesis H2 is not
73


supported by the data; that is Infrastructure resources (IR) on its own does not have a
significant influence on KMC (P = 0.037; t = 0.285; p > 0.05). The direct effect model also
was used to test the hypothesis H4 It is seen that KMC has a significant positive
influence on BPO Outcomes (P = 0.640; t = 5.045; p < 0.001).
The second-order factor model with the higher order construct, ISR is used to
test the remaining two hypothesis, H3 and H5. To estimate the hypothesized second
order formative model of ISR from the two first order factors of Social capital resources
(SCR) and Infrastructure resources (IR), the coefficients for each of the two first order
factors were modeled using a principal components factor analysis, following the
procedures in Karimi, Somers and Bhattacharjee (2007b) and Diamantopoulos and
Winklhofer (2001). The results of the factor analysis are shown in Tables 8.6 and 8.7.
74


Table 8.6
Principal Components Analysis for Second Order Model for ISR
Component
1 2
IR1 -.058 .933
IR2 .078 .932
PSN .871 .007
ID .810 -.018
GT .828 .037
Eigenvalue 2.115 1.736
% Variance Explained 42.304 34.718
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
Table 8.7 Component Score Coefficient Matrix for ISR
Component
1 2
IR1 -.038 .537
IR2 .026 .535
PSN .413 -.006
ID .384 -.019
GT .392 .012
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
75


The correlations among the two first order factors is very low, demonstrating that
the content captured by the first order factors are distinct from one another. The
coefficients of the first order factors are statistically significant, providing justification for
the existence of the hypothesized formative second order model of ISR.
It is seen that the second order model is a more parsimonious model with fewer
parameters to be estimated. The structural link from ISR to KMC (see Table 8.4) is
positive and significant (P = 0.732; t = 7.952; p < 0.001). The results provide empirical
support for H3 and validates the expectation that ISR as a set of resources has a
significant association with building KMC.
The predictive power of the models for the BPO outcomes construct is shown by the
R2 values in Table 8.4. The final hypothesis concerns the presence of an interaction between
ISR and KMC with BPO (H5), which states that ISR will moderate the relationships between
KMC and BPO. The interaction term (ISR x KMC) was formed by cross-multiplying all
standardized items of each constructs, following the procedure in Chin, Malcolm and
Newsted (2003) The results suggest an assembly of social capital resources and
infrastructure resources can slightly strengthen the influence building knowledge capabilities
has on business process outcomes (P = 0.654; t=3.670; p < 0.001) and supports H5.
Comparing the Moderator Model with Second Order Model
As shown in Figure 8.2, to assess the interaction of ISR and KMC on BPO the two
estimated models direct effect and second order models were compared in two steps to
judge the incremental variance explained by adding the moderating effect (Chin, Malcolm
and Newsted, 2003). The second order model was chosen to add the moderating effect to,
since the second order model produced a higher R-square for the BPO outcomes measure.
76


The R2 for BPO outcomes was 0.4275 for the second order model versus 0.4272 for the
direct effects model (see Table 8.4). The test for the moderated relationship was
conducted by using AR2 to draw conclusions about the moderator effect size since "the
use of the path coefficient of an interaction term will lead to spurious conclusions (Carte
and Russell, 2003). This approach prevents an incorrect interpretation of the
significance of the interaction term when it is correlated with its constituent parts, that is,
its main effect (Karimi, Somers and Bhattacharjee, 2007a). The change in R2 was
examined by comparing the results of two models derived from the second order model
(which was chosen as it has the higher R2 compared to the direct effect model See
Table 8.4). The R2for the BPO construct from these models are used to calculate the
moderating effect of ISR relationship between KMC and BPO. The relationship between
BPO and KMC can be specified as follows:
Base model: BPO = f (KMC, e...), (1)
and compared to the following model that includes the relationship of KMC and
the interaction of KMC and ISR:
BPO = f (KMC, KMC-ISR, £...), (2)
or more specifically by an equation1 of the form:
BPO = p0+PiWWC+p2 KMCHSR + E. (3)
1 The sensitivity of results to the inclusion of an interaction term is often taken as a sign of
multicollinearity. If there is high multicollinearity, it can lead to large standard errors on
the model parameters. However, and more importantly, we are not directly interested in
the significance or insignificance of the model parameters per se anyway. Instead, we are
interested in the marginal effect of X on Y. In the case of Eq. (3), this is: 8Y/5KMC = 0!
+ p2KMC*ISR (Karimi, Somers, Bhattacharjee, 2007a)
77


Consequently, AR2 denotes the significance of the interaction term when added
to the base model. The standardized path estimate from the product construct (ISR x
KMC) to BPO indicates how a change in ISR would change the influence of KMC on the
dependent construct BPO. The R-square for the second order model with the interaction
term (R2 = 0.4507) was compared with the R-square for the second order model without
the interaction term (R2 = 0.4275) to assess the strength of the moderating effect. The
true effect of the interaction term was calculated through the effect size2 (f2) where 0.02,
0.15, and 0.35 has been suggested as small, moderate, and large effects, respectively
(Cohen, 1988). The f statistics, which is based on the differences in R2 between the two
models, was determined and used to compute the pseudo-F statistic3 (Carte and Russell,
2003; Chin, Malcolm and Newsted, 2003). The interaction effect from ISR x KMC on
BPO produces a small effect size (f2= 0.023) thereby supporting the posited moderating
effects of ISR on the relationship between KMC and BPO. An F statistic that is
significantly greater than 1.00 leads to the rejection of H0: AR2 = 0. The Pseudo F-
statistic was calculated as F=4.519, thereby providing statistical significance to hypothesis,
h5.
The analysis suggests that the additional variance explained by introducing the
moderator, ISR X KMC significantly adds to the variance explained in BPO (R2 increasing
from .4275 to .4507 in the second order model when the moderator is included).
2 Effect size f2 = [R2 (included) R2 (excluded)] / [(1- R2 (included)]
3 The pseudo-F statistic is computed using the formula f (n-k-1), with 1, (n-k) degrees of
freedom where n is the sample size and k is the number of constructs in the model (Chin,
Malcolm and Newsted, 2003).
78


Consequently, it is evident that ISR and KMC demonstrate a mutually reinforcing
complementarity relationship (Zhu, 2004). ISR and KMC jointly improve BPO. Therefore,
the impact of KMC on BPO will be contingent on the level of IS resources; the impact will
be stronger for firms with the higher levels of IS resources in comparison to the firms with
the lower levels.
79


CHAPTER 9
DISCUSSION AND CONCLUSIONS
Three research questions were examined in this study: (1) Can IS resources -
particularly social capital resources from social capital theory translate to improved
knowledge management capabilities between the client and vendor in the offshore BPO
arrangement? (2) Do knowledge management capabilities impact the BPO outcomes?
(3) Do IS resources and knowledge management capabilities have a complementary
relationship with BPO outcomes?
To answer the questions, five research hypotheses were tested. Using survey
data from a field study of a technical support organization that deployed BPO, significant
evidence was found for the direct association of social capital resources (H,) and building
knowledge management capabilities, but not for the direct association of infrastructure
resources (H2) on knowledge management capabilities. From a practical perspective,
however, infrastructure resources are also important for both building knowledge
management capabilities and this is supported in the second order model where IS
resources (ISR) as a combination of social capital resources (SCR) and infrastructure
resources (IR) was significantly related to knowledge management capabilities (H3). The
model also supported the relationship between knowledge management capability on
BPO outcomes (H4). Finally the complementary effect of IS resources on BPO outcome
through KM capability building was also supported (H5). These findings demonstrate
that the Direct Effects model was not sufficient for explaining how firms build knowledge
management capabilities. There is an additive effect of the set of IS resources (ISR) on
building knowledge management capabilities where the effect of each resource
80


seemingly depends upon commingling with others. The study also found that the co-
presence of IS resources tends to supersede the direct effects of each IS resource alone.
Integrated together, they result in a performance-enhancing resource bundle for building
knowledge management capabilities. Therefore, the individual dimensions of IS
resources SCR and IR should not be considered in isolation from each other, but
should be treated in a collective and mutually reinforcing manner. This finding is similar to
those reported in recent studies (Karimi, Somers and Bhattacharjee, 2007a; Tanriverdi,
2005; Tanriverdi and Venkataraman, 2005). The co-presence of resources provides
unique value to the firm. Moreover, the synergies arising from the co-presence of
resources are much more difficult to observe and imitate by other firms (Tanriverdi,
2005).
Regarding the second question, the results did provide compelling evidence that
building knowledge management capabilities has positive association with BPO
outcomes. Also, the association of building knowledge management capabilities with
BPO outcomes is contingent on the co-presence of IS resources, which supports the third
research question as well. The importance of the co-presence of IS resources was
demonstrated by their addition to the model. When included, it increased the explained
variance in business process outcomes to 45 percent (R2= .4507), and had a small effect
size (f 2 of 0.0232). This confirmed the hypothesis that IS resources can intervene to
strengthen the relationship between building knowledge management capabilities and
business process outsourcing outcomes. The notion of complementarities suggests that
firms that possess a wide range of resources are better able to exploit the synergistic
benefits of these resources than those that possess fewer resources or a lesser level of
each resource (Tanriverdi and Venkatraman, 2005). This is important since competitors
81


usually lack the strategic foresight to recognize complementarities. Even, if they
recognize the complementarities, to imitate them successfully, they have to make
simultaneous changes in IT infrastructure (Karimi, Somers and Bhattacharjee, 2007a).
The client vendor relationship in a BPO can be quite arduous, with jobs at
stake and differences in capabilities of the two organizations. This study shows that if
social capital resource is increased in the BPO relationship, then the overall outcome of
the BPO can be improved. With the development of shared work practices, social capital
resource is created resulting in improved knowledge management processes. As a result
the BPO outcome is improved. Practitioners may note that the investment into building
successful working level relationships in a BPO can indeed be worthwhile.
Drawing on the resource based view of the firm, this research demonstrates that
IS resources particularly infrastructure and relationship resources can influence
knowledge management capabilities and improve BPO outcomes. The complementarity
effect of IS resources and knowledge management capabilities on BPO outcome implies
that investments into building knowledge management capabilities together with building
the IS resources in conjunction leads to higher BPO outcome gains.
Implications for Research
A research model based on RBV to study the contribution of social capital
resource as an IS resource on the knowledge management capabilities in a BPO is
developed and validated. By defining the social capital resource and measuring it in the
BPO setting this study makes a valuable contribution to the body of IS research. The
results of this study support the research background that benefits of social capital
resource can be seen in knowledge management capabilities. The results indicate that
the BPO outcome is improved. Researchers have been looking for the impact of social
82


capital resources in downstream work processes and this study shows that social capital
resources can improve outcomes by improving knowledge management capabilities. A
measurement of BPO outcome based on Coordination theory is also developed and
validated. This BPO measurement model may be used in future research to study other
BPO scenarios as well and measure the performance outcomes of a BPO. Additionally,
the study can be extended to other industries with knowledge intensive processes such
as the healthcare industry as more processes in that industry are outsourced (Scott and
Ghosh, 2005b).
Implications for Practice
The results support the need for building social capital resources in offshore BPO
and deploying IT infrastructure resources together with social capital resources to exploit
the complementarity effect. Several papers have studied the enablers for social capital
and management practices that can help foster outsourcing success as well as generate
social capital (Rottman and Lacity, 2004). The organizational enabling conditions stated
for supporting social capital creation fall into three categories: (1) Bridging, where
individuals are brought together purposely for collective work, (2) Bonding, where
cognitive norms and implicit understanding is developed by personnel on both sides, and
(3) Linking, where structural connections are established for jointly owning ongoing
activities (Kowch, 2005).
Organizational processes that involve resources on both sides of the relationship
can be social capital enablers, if such processes posses the above characteristics.
Examples of these organizational practices may be: (1) creating opportunities for
exchange, (2) creating an expectation that such combinations and exchanges will have
value, (3) creating motivation for both sides to participate and (4) creating structural
83


norms and symmetries to support combination capability (Nahapiet and Ghoshal, 1998).
Wenger (1998) found these factors and their inter-relatedness to significantly impact
social capital formation and knowledge transfer in communities of practice. Several
authors have argued that social capital creation is possible through pursuit of
organizational enabling activities (OBrien, Phillips and Patsiorkovsky, 2005; Schmid,
2000).
Delone, et. al. (2005) describes the differences between bridging, bonding and
linking activities in social capital creation. They argue that any mechanism that
establishes repetitive and routine aspects of tasks in the inter-organizational relationship
can be classified as bridging". Organizational activities that help organizations and
teams interact and coordinate implicitly by building inner expectations, through cognition
mechanisms of norms, beliefs and trust can be referred to as bonding activities.
Finally, Baum and Ziersch (2006) describe linking activities as organizational practices
aimed at building vertical connections, spanning difference of power structure to help
reduce inequalities and build common responsibility for tasks in the inter-organizational
relationship.
Generalizability of the Results
The data collection from a single organization can raise questions about the
generalizability of any theory. However it is usually recommended to replicate any study
in the new setting to justify porting any results or theory from the research setting to a
new setting (Lee and Baskerville, 2003). While the above approach to generalizing
results may be too restrictive, the authors go on to say that even if a 100 or 1000
organizations were included in the study, it would only extend generalizability to those
100 or 1000 organizations that fell in the scope of the study setting and not guarantee
84


universal generalizability. This undermines the need to include more than one
organization in a confirmatory research study to support the generalizability of the results
from the study.
The setting of this study was a large multinational customer services organization
that supports multi vendor networking equipment. The organization had involved an
offshore vendor organization to outsource some of the transactions in the service
processes. The processes by their description are fairly representative of the industry
and are more or less universally followed in delivering product support and network
design work through out the telecommunications and high technology industry by many
other organizations. Thus there are no peculiarities with these processes, or the
transactions that these processes support or the personnel involved in the processes.
Hence, there is no reason to believe that these results and the confirmed theory cannot
be applied to other customer service outsourcing settings or possibly other BPO settings
as well. Moreover, customer service and technical product support accounts for a major
percentage of offshore BPO at greater than 30% of all outsourced processes (Segal,
2003). Hence the results and the confirmed theory are of particular interest to the IS
community and do have general appeal.
Limitations
The limitations of this research are the fact that only one organization was
surveyed. Clearly these two processes for the selected organization may not be
representative of the general customer service processes that are being outsourced.
However, the descriptions of the processes do not indicate any particular peculiarities
that may render them so unique that other customer service processes in other
organizations would be grossly different.
85


The IS resources construct in the research model included relationship resources
in the form of social capital resources and infrastructure resources. A third form of IS
resource often used in RBV studies are knowledge resources (Karimi, Somers and
Bhattacharjee, 2007a). A firm's knowledge resources are the unique skills, expertise,
insights, experience, and intellectual capital that it uses for building a capability.
Knowledge exploitation requires sharing relevant knowledge among members of a firm to
promote mutual understanding and comprehension. While the types and forms of
knowledge resources for the organization were catalogued in the data collections section,
this study did not however, incorporate a measurement for knowledge resources in the
research model. This was a decision based on confusion regarding where such a
measurement may have fit in the model (1) in the outcomes measure as knowledge
resources could be considered units of coordination information or (2) in the resources
measure if knowledge resources were considered to be a resource for this study. It is a
fact that adequate and high quality knowledge resources are needed to facilitate the
effectiveness of knowledge management capabilities. If workers could not locate or
exploit these knowledge resources from the KMS, why would they increase their use of
the knowledge management capability? However, it can be safely assumed that there
were indeed adequate knowledge resources in the KMS. This was a precondition of the
research, since these service processes have been operational in the client organization
for many years using the same knowledge management systems and tools. The
processes were outsourced partially to an offshore vendor to increase staffing at a lower
labor rate. The knowledge resources that were existing in the KMS were still relevant
after the BPO implementation and hence form a resource base that is accessible by both
client and vendor staff. Given that the study has collected the detailed records of how
86


many knowledge entries were added over a 8 month duration in the various knowledge
management tools, it is also quite feasible to develop a measure of knowledge resources
and incorporate into the model and retest the hypotheses for any changes in the results
as a future exercise.
87


APPENDIX A FINAL SURVEY
QUESTIONS Strongly Disagree Disagree -2 Neutral 3 1
Agree 4 Strongly Agree -5
1. What Outsourced process are you answering the questions with respect to? a. Customer Support b. Network Consulting
2. Years of on the job experience?
3. Years of schooling (K-12 counts as 13years)
4. Gender? a. Male
b. Female
There is a norm of teamwork among staff in the BPO relationship. 1 2 3 4 5
There is a norm of openness of diverse/conflicting views among staff in the BPO relationship 1 2 3 4 5
There is a norm of tolerance of mistakes among staff in the BPO relationship 1 2 3 4 5
Client and vendor staff in the BPO relationship share common values. 1 2 3 4 5
Client and vendor staff in the BPO relationship share common goals 1 2 3 4 5
Client and vendor staff in the BPO relationship have strong group identification 1 2 3 4 5
Client and Vendor staff in the BPO relationship give credit to people where it is due 1 2 3 4 5
Obligations between organizations are always sustained by staff in the BPO relationship 1 2 3 4 5
Client and Vendor staff in the BPO relationship use other's knowledge appropriately 1 2 3 4 5
Our organizations have processes for acquiring knowledge from our BPO partners 1 2 3 4 5
Our organizations have processes for generating new knowledge from existing knowledge in the BPO relationship 1 2 3 4 5
Our organizations have processes for exchanging knowledge between individuals in the BPO relationship. 1 2 3 4 5
Our organizations have processes for converting knowledge into the design of future BPO support services 1 2 3 4 5
Our organizations have processes for distributing knowledge throughout the two organizations in the BPO relationship 1 2 3 4 5
Our organizations have processes for integrating different sources and types of knowledge 1 2 3 4 5
88


Our organizations have processes for applying knowledge learned from experience 1 2 3 4 5
Our organizations have processes to make knowledge accessible to those who need it 1 2 3 4 5
Our organizations have processes to quickly link sources of knowledge in solving problems 1 2 3 4 5
1 am encouraged to discuss my work with employees in other workgroups 1 2 3 4 5
Benefits of sharing knowledge in my organization outweigh the costs 1 2 3 4 5
Knowledge is freely exchanged freely through email, presentations, newsletters, etc. 1 2 3 4 5
Our organizations have processes to protect knowledge from inappropriate access in the BPO relationship 1 2 3 4 5
Our organizations have processes to protect knowledge from inappropriate use in the BPO relationship 1 2 3 4 5
Our organizations clearly communicate the importance of protecting knowledge 1 2 3 4 5
After the BPO, the my organization adapts more quickly to unanticipated changes 1 2 3 4 5
After the BPO, the my organization can react more quickly to new market info 1 2 3 4 5
After the BPO, my organizations response to market changes involves a less lengthy process 1 2 3 4 5
After the BPO, the transaction costs related to these outsourced processes are lower 1 2 3 4 5
After the BPO, we can serve our customers more effectively now 1 2 3 4 5
After the BPO, less resources are required to operate the outsourced processes 1 2 3 4 5
After the BPO, a decreased number of resources are needed to support the outsourced process 1 2 3 4 5
After the BPO, a decreased number of managerial contacts are needed to support the outsourced process 1 2 3 4 5
After the BPO, less time is required to get status and updates about the outsourced process 1 2 3 4 5
Appropriate hardware, software, and network infrastructures are in place to support the BPO 1 2 3 4 5
Appropriate knowledge systems and tools are in place in the infrastructure to support the BPO 1 2 3 4 5
Access technologies are administered in the infrastructure to allow appropriate access to knowledge 1 2 3 4 5
89


Full Text

PAGE 1

THE ROLE OF SOCIAL CAPITAL IN DETERMINING BPO OUTCOMES by Biswadip Ghosh B.Tech., Indian Institute of Technology, 1986 M.S., Colorado State University, 1988 M.B.A., Regis University, 1993 A thesis submitted to the University of Colorado at Denver and Health Sciences Center in partial fulfillment of the requirements for the degree of Doctor of Philosophy Computer Science and Information Systems 2007

PAGE 2

This thesis for the Doctor of Philosophy degree by Biswadip Ghosh has been approved by Ellen Stevens Date

PAGE 3

Ghosh, Biswadip (Ph.D., Computer Science and Information Systems) The Role of Social Capital in Determining BPO Outcomes Thesis directed by Assistant Professor Judy E. Scott ABSTRACT Organizations are pursuing the outsourcing of business processes (BPO) to offshore locations. However, current research has shown that knowledge management issues between the client and vendor organizations leads to less than expected benefits in many BPO arrangements. This research applies the Resource Based View (RBV) framework to study the impact of Social Capital resources on building Knowledge Management Capabilities in the BPO as well as the ultimate BPO outcome. The essence of RBV theory is that when investments are made into building IS resources, the impact of those investments on the firm's performance manifest through the firm's capabilities. Complementarity theory also suggests that a resource produces greater returns in the presence of another resource or capability than it can produce by itself. Firms pursuing BPO may be at different stages of the deployment of knowledge management capabilities and the IS resources of infrastructure and social capital and therefore the capabilities of the engagement may not be being fully exploited. Hence the impact of social capital and infrastructure resources on knowledge management capabilities and/or their combined impact on BPO outcome could explain the variability seen in BPO outcomes. A measure of BPO outcome based on the three information flow costs from coordination theoryvulnerability, production and coordination is also developed. A field study of a knowledge management system (KMS) in a BPO is used to test the hypotheses. Results indicate that increasing social capital resources alone and as well as in combination with increasing infrastructure resources in the client-vendor BPO

PAGE 4

infrastructure resources in the client-vendor BPO arrangement leads to an increase in management capabilities. However, increasing infrastructure resources alone has no significant effect on building knowledge management capabilities. Moreover, the increase in the knowledge management capabilities also significantly increased BPO outcome. Finally, this study also confirmed that IS resources can intervene to strengthen the relationship between building knowledge management capabilities and business process outsourcing outcomes as a small complementarity effect was supported. This has an important implication for practitioners, who seek to improve BPO outcomes that capabilities in the BPO arrangement must be developed in conjunction with the deployment of IS resources. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. (J (jJdy E. Scott

PAGE 5

DEDICATION I dedicate this thesis to my parents, Jaya and Bireswar Ghosh, who gave me an appreciation of learning and taught me the value of perseverance and resolve. I further dedicate this to my wife, Dalia, for her unfaltering support and considerable understanding while I was completing this work. I also dedicate this thesis to my two children, Juhi and Avishek for their acceptance of the many hours I spent in my home office, often forgetting them, as they stood in front of the French doors of the study trying to get my attention and a moment of my time.

PAGE 6

ACKNOWLEDGMENT My thanks to my advisor, Dr. Judy E. Scott, for her contribution and support to my research. I wish to thank Dr. Jahangir Karimi for his considerable input and guidance. I also wish to thank all the members of my committee for their valuable participation, review, recommendations and insights into this work and its final form.

PAGE 7

TABLE OF CONTENTS Figures ................................................................................................................... x Tables ...................................................................................................................... xi CHAPTER 1. INTRODUCTION .................................................................................................. 1 Resource Based View .................................................................................. 3 Information Systems Resources .................................................................. 4 IT Infrastructure Resources ...................................................................... 5 Social Capital or Relationship Resources ................................................ 6 Knowledge Resources ............................................................................. 8 Knowledge Management Capabilities .......................................................... 8 Research Questions ................................................................................... 14 2. OUTSOURCING RESEARCH BACKGROUND ............................................... 15 Outsourcing Research ............................................................................... 15 Outsourcing Relationship Research .......................................................... 17 Outsourcing Outcomes Research .............................................................. 19 3. THEORETICAL FOUNDATIONS ....................................................................... 21 Social Capital Theory ................................................................................. 22 BPO Outcomes .......................................................................................... 30 Coordination Theory ................................................................................... 31 4. DEFINING STUDY CONSTRUCTS .................................................................... 33 Social Capital Resources ........................................................................... 33 Group Identification ................................................................................ 34 Generalized Trust. .................................................................................. 34 vii

PAGE 8

Pro Sharing Norms ................................................................................. 35 Infrastructure Resources ............................................................................ 36 Knowledge Management Capabilities ........................................................ 36 BPO Outcomes .......................................................................................... 37 5. RESEARCH HYPOTHESES ............................................................................. 41 6. RESEARCH METHODOLOGY .......................................................................... 50 Research Design ........................................................................................ 50 Study Variables .......................................................................................... 50 Measurement of Research Variables ......................................................... 51 7. DATA COLLECTION .......................................................................................... 53 Collection Methodology .............................................................................. 53 Organizational Practices for Knowledge Management.. ............................ 53 Outsourced Processes ............................................................................... 55 Customer Service Process ..................................................................... 55 Network Consulting Process .................................................................. 56 Survey Results ........................................................................................... 57 Knowledge Resources ............................................................................... 61 8. DATA ANALYSIS AND RESULTS ..................................................................... 62 Estimation of Internal Consistency ............................................................. 68 Dimensionality, Convergent and Discriminant Validity .............................. 70 Second Order Model .................................................................................. 72 Hypothesis Testing ..................................................................................... 73 Comparing Moderator Model with Second Order Model ........................... 76 9. DISCUSSION AND CONCLUSIONS ................................................................. 80 Implications for Research ........................................................................... 82 Implications for Practice ............................................................................. 83 viii

PAGE 9

Generalizability of the Results ................................................................... 84 Limitations .................................................................................................. 85 APPENDIX A. FINAL SURVEY ................................................................................................ 88 B. OUTER MODEL LOADINGS (DIRECT EFFECTS MODEL) ........................... 90 C. OUTER MODEL LOADINGS (SECOND ORDER MODEL) ............................. 93 BIBLIOGRAPHY .......................................................................................................... 96 ix

PAGE 10

LIST OF FIGURES Figure 1 .1 Research Background ............................................................................................. 13 3.1 Theoretical Foundations .......................................................................................... 29 5.1 Research Model ...................................................................................................... 41 8.1 Direct Effects Model ................................................................................................ 63 8.2 Second Order Model ............................................................................................... 64 X

PAGE 11

LIST OF TABLES Table 4.1 Survey Instruments ................................................................................................. 39 7.1 Demographics across Processes ........................................................................... 59 7.2 Descriptive Statistics of Items ................................................................................. 60 8.1 Correlations, CR and AVE for Direct Effects Model ................................................ 66 8.2 Correlations, CR and AVE for Second Order Model ............................................... 67 8.3 Internal Consistency Measures ............................................................................... 69 8.4 Structural Model Results ......................................................................................... 71 8.5 Correlation Matrix of Factors in ISR Second Order Model ..................................... 72 8.6 Principal Components Analysis for Second Order Model for ISR ........................... 75 8.7 Component Score Coefficient Matrix for ISR .......................................................... 75 xi

PAGE 12

CHAPTER 1 INTRODUCTION The concept of transferring the operational ownership and execution of one or more business processes is referred to as Business Process Outsourcing (BPO). In an offshore BPO, the vendor organization is located in a different country from the client organization, that is seeking to outsource. Due to the rapid globalization of information technology and improvements in telecommunications networks, firms are pursuing offshore outsourcing to remote countries like India and China. Offshore outsourcing allows the domestic firms to tap into large pools of educated workers at significantly lower labor costs. Additionally, client firms benefit by getting access to large markets for products and services in the foreign countries and take advantage of time zone differences to staff round the clock operations using a "follow the sun" policy (Lieberman, 2004). The vendor organization undertakes responsibilities of the execution of the process. Examples of BPO are seen in several industries such as finance, healthcare and technical support. Both core and supporting (non core) processes in a client firm's value chain are being outsourced to offshore vendors. The offshore BPO market is the fastest growing segment of the overall IT outsourcing market and is projected to grow 60% year over year (Trapper, 2003). However, risks in offshore outsourcing are significant and can lead to less than expected benefits (Aron, Clemons and Reddi, 2005; Scott and Ghosh, 2004). Actual outsourcing case studies indicate that many organizational goals remain unfulfilled in these arrangements (Lacity and Willcocks, 1998). Their study reported that only 54% of the outsourcing agreements that they surveyed realized cost savings. These numbers were recently confirmed by a practitioner study by (Moore, 2004; Moore and Martorelli, 2004). Such variability in the

PAGE 13

BPO outcomes call for empirical research into the drivers for success in such BPO arrangements. Current BPO research literature points out the need for bi-directional knowledge management capabilities in offshore BPO arrangements. While the risks and challenges are significant in an offshore outsourcing relationship, knowledge management practices have been found to play a major role to improve the effectiveness of these arrangements (Willcocks, Lacity and Kern, 1999). Using case studies, Rottman and Lacity (2004) identified 20 best practices that can be used to overcome the difficulties. While this study did not specifically study KM practices, yet KM initiatives such as balanced scorecard metrics, real time dashboards and identifying subject matter experts were part of the list of 20. A study of knowledge management practices in IT outsourcing by Willcocks, et. al. (2004) indicated the importance of social capital resources in outsourcing relationships. Another case study of the organizational impact of the outsourcing decision at Logistics Information Systems Agency (LISA) presented by Willcocks, Lacity and Kern (1999) found several risks that are not fully considered when the BPO was implemented, but later on needed management resources to alleviate. These issues are pronounced in the high technology industry, where knowledge based processes are being outsourced. High technology firms differ from other organizations in that they operate in a more dynamic and innovative environment with the need for efficient information flow. Existing research studies on outsourcing and associated organizational factors have to date ignored the specific characteristics of the BPO for such knowledge based processes (Lacity and Willcocks, 1995, Levina and Ross, 2003, Carmel and Agrawal, 2002). An important question that has not been studied in IS research is how relational and 2

PAGE 14

infrastructure resources contribute to the BPO outcome for knowledge intensive processes in the high technology industry. As more and more knowledge intensive work is moved offshore, the need for bi-directional knowledge management capabilities in the BPO arrangement is increasing particularly for professional and high technology work processes (Ghosh and Scott, 2005d; Ghosh and Scott, 2006a). Resource Based View The Resource Based View (RBV) of the firm states that "resources are essential raw materials for capability-building and their availability determines the firm's ability to build such capabilities, which are often critical drivers of firm performance" (Wade and Hulland, 2004; Barney, Wright and Ketchen, 2001; Bharadwaj, 2001). Capabilities are defined as repeatable patterns of actions in the use of those resources to create, produce and offer products/services to the market. Capabilities can include management ability and skills and processes/systems that allow for creation, storing and sharing of knowledge (Wade and Hulland, 2004). Resources can be independently valued and traded, while capabilities cannot. The RBV framework supports modeling complementary relationships between resources and capabilities to determine whether they have a complementary impact on outcomes (Ray, Muhanna and Barney, 2005). The outcomes are defined as the unit of the firm's performance. Complementarity arises when a resource produces greater returns in the presence of another resource than it produces by itself. While the resource-based view (RBV) of the firm recognizes the complementary role of resources, it is not well developed in the business process outsourcing research and theory. However, studies in inter organizational alliances and partnerships have applied RBV and found that it is a suitable theoretical foundation to explain the outcome of the alliance and partnerships (Lavie, 2006). BPO falls in the 3

PAGE 15

domain of strategic networks (Gulati, Nohria and Zaheer, 2000) and therefore, RBV is a suitable framework to apply to study these inter organizational arrangements. Information Systems Resources One of the benefits of applying RBV theory is that several prior studies had categorized and defined Information systems resources, which provides a foundation for incorporating applicable IS resources into this study. The RBV literature provides several different categorization of Information Systems resources that have been used in RBV research. Ross, Beath and Goodhue (1996) identified technical, human, and relationship resources as three categories of IS resource. Another categorization used by Bharadwaj (2000) consists of tangible, human, and intangible resources. The common theme among the above categorizations fall into two resource areas tangible infrastructure resources and intangible human relationship management resources. Wade and Hulland (2004) listed 8 categories of Information Systems resources -(1) Managing external relationships, (2) Market Responsiveness, (3) Managing internal relationships, (4) IS planning and change management, (5) IS Infrastructure, (6) IS technical skills, (7) IS development, and (B) Cost Effective IS Operations. They also recommend that for research, which examines specific technologies, models using a set of more narrowly defined resources is appropriate. Feeney and Willcocks (1998) framed the outsourcing question on IS resources and recommend that non core IS resources be outsourced by the firm. They identify 9 IS resources -(1) Business Systems Thinking, (2) Relationship building, (3) Leadership, (4) Contract Facilitation, (5) Vendor Development, (6) Contract Monitoring, (7) Architecture Planning, (8) Informed buying and (9) Making Technology work. Arguably the first six resources can be seen as relationship and people based, while the last 3 account for the physical infrastructure. 4

PAGE 16

As seen from the previous section, two broad classes of IS resources need be considered for this study-(1) relationship resources and (2) infrastructure resources. These resources are heterogeneously distributed across the two firms in the BPO and impact operations on either side, and their presence (or absence) and complementarities with knowledge management capabilities may explain the variability of BPO outcomes. IT Infrastructure Resources Infrastructure resources refer to a firm's shared IT assets (e.g., hardware, software tools, and networks, databases and data centers). They are the foundations for a firm's IT architecture, which is the blueprint to support multiple business processes and user groups across the two firms in the BPO arrangement. Reliable IT infrastructure resources will ensure the success of IT architecture, which tends to be highly firm specific and evolves over a long period of time during which gradual enhancements are made to reflect changing business needs (Karimi, Somers and Bhattacharjee, 2007a). As the infrastructure becomes distributed throughout firms and even beyond their boundaries in a BPO, the firms need a clear vision of where to locate individual technology components and responsibility for those components. Firms with a valuable technology asset are developing architectures that elaborate rules for distributing hardware, software, and support as well as controlling their accessibility to user groups based in multiple firms -independent of individual applications. These rules specify what kinds of data to share and how to store them, where to locate servers, and how to support applications and technologies (Feeney and Willcocks, 1996). Firms without well-defined architectures have dealt with the challenges of distributed computing by first delivering systems and then thinking about how to connect and support access to them. The result is that their systems are either poorly supported or expensive to operate or both. In firms with a well5

PAGE 17

defined architecture, infrastructure integration requirements and support considerations drive system design so that new systems not only meet business needs but are also cost effective. Social Capital or Relationship Resources Social capital (relationship) resources consists of sharing ownership, risk and responsibility of the firm's business operations across the client and vendor personnel (Ross, Beath and Goodhue, 1996). Ross, Beath and Goodhue (1996) found that close working relationships across firms allow staff to observe business processes in action and accumulate experience in solving real business problems. Enabling interactions is the central foundation of social capital theory (SCT). Nahapiet and Ghoshal (1998) define social capital resource as '"the sum of the actual and potential resources embedded within, available through, and derived from the network of relationships possessed by an individual or social unit. Social capital thus comprises both the network and the assets that may be mobilized through that network". SCT theory argues that effective knowledge combination, exchange and transfer occur when sufficient opportunities for interactions and exchange are present in the inter organizational relationship between individual knowledge workers (Nahapiet and Ghoshal, 1998). Social capital is defined in three dimensions with each dimension contributing to the meaning of social capital and where each dimension alone is not able to fully capture the concept in its entirety. The three components are commonly seen as: (1) Generalized Trust or the relational aspects along with types/tones of social interactions, (2) Identification or the issues that relate to structural identification with the cross firm work group, and (3) Pro Sharing Norms or the cognitive issues of rules and norms governing social action (Coleman, 1988, Putnam, 1993, Nahapiet and Ghoshal, 6

PAGE 18

1998). Social capital resources only exist within a relationship and cannot be separated from the network. They are created through exchanges that happen when the parties in the relationship facilitate opportunities for interaction (Ghosh and Scott, 2007b). In a large empirical study of public organizations in Singapore, components of social capital such as organizational norms and structure were found to moderate the usage of electronic knowledge repositories in managing knowledge (Kankanhalli, Tan and Wei, 2005). Given these results, it is clear that social capital resources in the relationship between the client and vendor personnel is a key concept in BPO arrangements that need to be explored further. Social capital resources form an important complementary resource to the infrastructure resources, which have traditionally been deployed to manage outsourcing projects in the IT and IS domains (Rottman and Lacity, 2004; Delane, et. al., 2005). Such social capital resources have the potential to mitigate risks and address unaccounted for scenarios that are not specified in the traditional work contracts. Social capital resources have also been found to be an important component in the successful adoption and use of knowledge management systems (Wasko and Faraj, 2005; Huysman and Wulf, 2006). The research on inter-organizational knowledge management also shows that knowledge management between organizations is affected by four factors that include the characteristics of the organizations and their relationship (Argote, 1999). Ko, Kirsch and King (2005) reported that the more arduous the relationship, the less knowledge transferred, while greater the shared understanding, credibility and motivation, greater the knowledge transferred. lnkpen and Tsang (2005) also found that social factors play a significant role in knowledge management in social networks. The stream of research in partnerships and inter-organizational learning has 7

PAGE 19

also identified that social capital in relationships play a role in knowledge transfer in both formal and informal settings. Burges (2005) found both individual relational resources and organizational infrastructure resources impacted knowledge management across firms. Similar results in the domain of inter-organizational relationships can be noted from Scott (1998), Malhotra, Gossain and El Sawy (2005) and Argote (1999). Knowledge Resources A third category of IS resources identified in RBV literature are the knowledge resources (Karimi, Somers and Bhattacharjee, 2007a). Knowledge resources are the knowledge assets that are in the knowledge management system. These knowledge assets or resources are the artifacts of the BPO arrangement. These knowledge artifacts fall into three categories product knowledge, customer knowledge and managerial knowledge (Tanriverdi and Venkatraman, 2005). Product knowledge refers to research and development and operations knowledge by which the firm develops and produces its products and services. Customer knowledge refers to needs, preferences and intelligence about customers and markets of the firm, specifically in the areas of marketing and advertising skills and policies. Managerial knowledge refers to the knowledge required for governing the business units of the firm and include managerial practices, policies and processes of the firm (Grant, 1991 ). Knowledge Management Capabilities Cross unit knowledge management capabilities are important sources of synergies in multi-firm arrangements such as BPO (Tanriverdi, 2005). These capabilities require coordination processes across the firms as well as knowledge assets or artifacts. Since knowledge has a personalization aspect to it, hence it is more difficult to manage and manipulate than mere information. Knowledge management capabilities are defined as 8

PAGE 20

organizational processes that allow the firms to create, exoloit and protect these knowledge assets (Alavi and Liedner, 2001; Davenport and Prusak 1998). Gold, Malhotra and Segars (2001) identified a set of knowledge management processes that an organization utilizes to transfer and exploit knowledge from external sources, such as partnerships. They mentioned that these processes require the presence of social capital resources and suitable IT infrastructure resources deployed in the firm. They identified four dimensions acquisition, conversion, application and protection. Zahra and George (2002) identified these processes as a reconceptualization of the absorptive capacity of a firm, as dynamic capabilities, which four underlying organizational processes which are each comprised of specific routines and procedures that allow the identification of knowledge and its exploitation. They list absorptive capacity in knowledge management as either (1) "potential capacity", which consists of acquisition and assimilation capabilities or (2) "realized capacity", which are the transformation and exploitation capabilities. These capabilities can exist across firms by the institution of a set of organizational processes that lets the firms coordinate knowledge management activities. There are five interrelated processes that are critical for managing cross-unit knowledge synergy: (1) Creation or Acquisition of Knowledge (Nonaka, 1994; Gold, Malhotra and Segars, 2001), (2) Integration or Conversion of knowledge ((Grant, 1996), (3) Leverage or application of knowledge (Gold, Malhotra and Segars, 2001) and (4) Transfer of knowledge (Argote, 2001, Szulanski, 1996). A final component of knowledge management capability is the (5) protection of knowledge (Gold, Malhotra and Segars, 2001 ), which is particularly important in offshore BPO arrangements, where intellectual capital abuse is quite common and well documented in 9

PAGE 21

practitioner literature. This research stream also suggests that if any one of these capabilities is low, then firm performance can be reduced (Tanriverdi, 2005). Hence, knowledge management capabilities are a set of capabilities that provide a reflection of the firm's overall capacity to exploit knowledge in its business. The current research notes the need for knowledge resources that are relevant and applicable across multiple business units is essential for cross unit knowledge synergies. Those resources need to be transferred and integrated with business operations for the firm to realize value. Finally the resources must be leveraged by changing the behavior of the firm in order to achieve the expected performance benefits. These knowledge capabilities provide cross unit synergies and mutually support each other (Venkatraman and Tanriverdi, 2005). Research specifically aimed at studying the impact of social capital resources on the outcome of offshore BPO relationships, while being well motivated from the IS literature, is still lacking. Such social capital resources are seen to impact knowledge management capabilities in prior research, while knowledge management capabilities are seen to improve inter organizational outcomes. Hence, a study to model the impact of social capital resource on the knowledge management capabilities in a BPO and the ultimate outcome of a BPO holds promise. Such a study has not been attempted in the IS research literature, and holds significant promise to explain the ambivalent results of BPO (Lacity and Wilcocks, 1998). Additionally, researchers of SCT have called on research into the diffusion and exploitation of the social capital resources in downstream knowledge management capabilities in the organization, as well as measuring the impact of SCT using transaction and process costs (Nahapiet and Ghoshal, 1998). While social theories can be used to 10

PAGE 22

ascertain benefits in inter-organizational knowledge management, information flow based cost measures are needed to get a better unbiased measure of outcomes. Such a measurement of BPO outcome using information costs is also lacking in the IS research stream. Coordination theory is based on the coordination, production and information costs associated with the interactions of organizational subunits (Malone and Crowston, 1999). A measurement model based on coordination theory can shed light on the effective flow of information and knowledge and the allocation of resources, both of which are critical components of any business process (Malone and Crowston, 1994) and therefore are critical to a BPO. Hence, using coordination theory to evaluate the outcome of offshore BPO holds promise. This research posits that high technology firms that invest in IS resources relational and infrastructure to facilitate knowledge management capabilities to support the BPO may see improved BPO outcomes through the impact of those investments in resources on the capabilities in the inter organizational partnership. Indeed outsourcing success is significantly linked to factors that fall beyond the written contract between the client and vendor (Koh, Ang and Straub, 2004). These include IT infrastructure resources and the development of relationship/social capital in the form of human resources within the outsourcing teams. Complex business obligations can never be completely written up. Koh, Ang and Straub (2004) stress the need to have psychological contracts that can drive the behavior of vendor personnel. The focus is on the individual level rather than at the organizational level. This individual focus is consistent with the working of knowledge processes in professional organizations such as technical support and healthcare (Ghosh and Scott, 2005b, Ghosh and Scott, 2005c, Ghosh and Scott, 200Gb). Often the intent of complex BPO contracts are lost among the 11

PAGE 23

vendor employees and leads to over specification. Improved BPO success can be achieved through relationship building at the individual level to support the established IS infrastructure resources that have already been deployed (Hislop, 2002) However, neither the impact of social capital on knowledge management capabilities in a BPO nor its impact on BPO outcome have been studied in IS literature This research poses the following fundamental question: Can Information Systems (IS) resources consisting of: (1) Social Capital resources from social capital theory and (2) Infrastructure resources facilitate knowledge management capabilities in a BPO? This study postulates that greater IS resources in the BPO relationship can directly lead to improved knowledge management capabilities which results in improved BPO outcomes Moreover the increased IS resources in combination with the increased knowledge management capabilities also results in improved BPO outcomes, beyond that produced by the capabilities alone. In summary, from a theoretical perspective, this study aims to investigate the role of social capital and infrastructure resources in offshore BPO arrangements by measuring the impact of those IS resources through its facilitation of knowledge management capabilities using the resource based view (RBV) of the firm This study is motivated by the BPO and social capital literature and holds promise of explaining the variability in observed BPO results, as well as providing a validated model to measure the BPO outcome in future BPO research. The Figure 1 1 summarizes the above research literature review in this conceptual domain 12

PAGE 24

Ghoshal (1998) Ro. Klrsch. K1ng (2005) Huysman. Wulf (2006 ) Argote (1999) Wasko, Far'\) (2005) Wade, Hulland (2004 ) Feeney, Willcock.s (1998) Rarum, Somers, Bhatlacharya (2007a) Figure 1.1 Research Background Lacity, Willcocks (1998) Wulcock.s. et. al (2004) Ghosh. Scott (2005 2006b) Cohen. Levintbal (1 990) Andreu Ciborra (1999) Huber (1991), Scott(1998) Daveop o rt. Prusak (1998) B u siness Process Outsourcing Outcomes (BPO) Koh, Ang Straub (2004) Rottman, Lacity (2004) DeLone, et. al. (2005) Coprpr:ano, Mitchell (2005) W!llcock.s, Lacuy, Kern (1999) Hislop (2002) Karimi, Somers, Bhattacharya (20 07a) 13

PAGE 25

Research Questions The goals of this research project are to answer the following questions using the RBV framework to model the deployment of IS resources to support inter firm capabilities in an offshore BPO arrangement: 1. Can IS resources particularly social capital (from social capital theory) resources translate to improved knowledge management capabilities between the client and vendor in the offshore BPO arrangement? 2. Do knowledge management capabilities positively impact the BPO outcome? 3. Do IS resources and knowledge management capabilities have a complementary relationship with BPO outcome? 14

PAGE 26

CHAPTER 2 OUTSOURCING RESEARCH BACKGROUND Outsourcing Research The research stream in outsourcing shows considerable diversity in four dimensions -(1) the stage of the BPO ranging from planning to implementation, (2) the organizational level where the study was conducted, (3) the theory applied and (4) the methods used. IS researchers have investigated different stages in the outsourcing decision process starting with the initial outsourcing decision to the outcome of the decision and have applied various theories and methods at different levels of analysis from the industry to individual workers (Dibbern, et. al., 2004). The staged decision making model separates the process of making the decision to outsource into four stages -(1) why outsource, (2) what to outsource, (3) how to manage the outsourcing implementation, and (4) finally measuring the outcomes of outsourcing. Several studies have investigated the questions of why a firm decides to outsource and what processes or systems they choose to include in the outsourcing arrangements (Loh and Venkatraman, 1992; Lacity and Willcocks, 1995; Hu, Saunders and Gebelt, 1997). These studies have found that the determinants of the outsourcing decision comes from both inside and outside the firm. The impetus for the decision to outsource is primarily strategic. Factors such as the firm's desire to reduce costs and their need to increase operational focus have been found to be statistically significant in this decision stage (Ang and Straub, 1998). Likewise, external industry trends and influences have been found to be significant as well in the decision to outsource (Loh and Venkatraman, 1992). These external influences include the so-called "Kodak effect", where the firm 's decision to outsource is significantly influenced by the decision of other 15

PAGE 27

firms in their industry as well as industry best practices and edicts from industry consultants. The content and selection of the processes to outsource, however, has been found to be a decision primarily driven by internal aspects of the firm and with lesser influence of industry and external factors. Both strategic and tactical characteristics of the firm have been found to be significant in the choice of processes to outsource (Lacity and Willcocks, 1995; Lacity and Willcocks, 1998; Ang and Slaughter, 1998). The choice can be a strategic one taken at the firm level, however it has influences from the tactical level in terms of political, structural and cultural characteristics of the firm as well as the profile of the processes themselves. The nature of outsourcing is defined by the use of a single or multiple vendors, while the degree of outsourcing is defined as the percentage of the function being outsourced up to complete outsourcing. Both of these characteristics can be determined by the strategy and culture of the firm(s) under study as well as the best practices in the firm's industry. Additionally, the type of outsourcing general versus transactional versus functional (business process outsourcing) is also a determinant of the why and what decisions (Hu, Saunders and Gebelt, 1997). The above research papers studies have utilized various theories including transaction cost theory, game theory, resource theory and other strategic management theories, and various relationship theories including social exchange theories. The selection of the theory to use is coupled with the level of analysis, such that resource based theories have been used at the organizational level to study internal determinants of the firm's resources and capabilities (Barney, 1991 ), while relationship based theories 16

PAGE 28

have been utilized to study cooperation and interactions at the individual level (Kern, 1997). Outsourcing Relationship Research The second group of outsourcing studies have focused on the implementation aspects of the outsourcing decisions to delve into the question of "how". These studies have investigated issues in the selection of the vendor and the building, structuring and management of the outsourcing relationship between the client and vendor (Kern, 1997; Grover, Cheon and Teng, 1996). Industry and firm level factors have been identified to be significant in the selection of the outsourcing vendor. While written contracts and service level agreements are important monitoring aspects of the outsourcing relationship, firms have recognized that psychological relationship management is an important part of managing the ongoing outsourcing contract (Koh, Ang and Straub, 2004). Various firm level characteristics have been found to have a significant influence in the outsourcing relationship trust, cooperation, social and cultural bonds, shared vision, expectations and norms as well as commitment (Kern, 1997; Willcocks and Kern, 1998; Grover, Cheon and Teng, 1996). Outsourcing arrangements come in varied time duration-short to long term; and contractual arrangements from a loose 'pay for service' arrangement to stronger partnerships such as joint ventures. Studies have also found relationships form a link between how an outsourcing project is structured and how it is managed, however, the length of the association has no direct impact on the relationship between client and vendor. For more strategic partnerships or alliances, the need for flexible governance mechanisms become more important. In such scenarios, the overall conclusion has been that contracts are necessary, but not sufficient for outsourcing success, rather 17

PAGE 29

relationship building becomes more important as the outsourcing becomes increasingly strategic for the firm (Clark, Zmud and McGray, 1995). The authors have defined and used several constructs to measure relationships to include such constructs as trust, communications that include adhoc channels, cooperation, etc. (Grover, Cheon and Teng, 1996). Various social and relational exchange theories and contract theories have been utilized in these studies to investigate process and management issues in outsourcing relationships between the client and vendor. Case studies and other empirical approaches using both positivist variance based models as well as interpretist descriptive methods have been used in this stream of research. Kern (1997) built a model that extended the structural contract theory models to include relational determinants. These studies placed their level of analysis at the worker level, which was found to be appropriate to evaluate relationships between client and vendor (Willcocks and Kern, 1998). Other studies have been done at the individual level and have focused on the work processes of outsourced IS professionals and the nature of the process/function being outsourced (Schultze and Boland, 2000). They found that the nature of work practices can jeopardize the relationship aspects in the outsourcing when the relationship is asymmetrical with weak group identification and vendors are setup as "fall guys" for potential organizational failures. In another study at the individual/functional level, Sabherwal (1999) found that outsourced projects proceed through cycles that involve trust, structure and performance. Moreover, when a balance is established between contractual structure and relationships the client firm is better able to manage changes and resolve unanticipated problems through BPO management. 18

PAGE 30

Thus it is evident from the above that studies that focus on vendor selection and contractual provisions are well suited at the firm level, while investigations into the relationship issues in a BPO can be more illustrative at the functional or individual level. Outsourcing Outcomes Research A final group of outsourcing research has focused on evaluating the outcome of outsourcing. These papers have studied the outcomes or success using constructs that are based on the realization of the expectations of the outsourcing from the firm's perspective (Lee and Kim, 1999; Aubert, et. al., 1999). Measures used in the studies include such constructs as cost savings, satisfaction, risk avoidance and relationship quality, such as conflicts at the firm level. Other authors have defined and used relationship constructs at the individual level such as fulfillment of responsibilities, treatment of workers, performance differentials and trustworthiness in the relationship (Ang and Slaughter, 1998). Various theories have been used to develop measures of BPO outcome. Transaction cost theories (TCT) were used in a seminal study of outsourcing outcomes by Lacity and Willcocks (1995). While TCT can be utilized to evaluate certain types of outsourcing contracts that are deemed classical with limited emphasis on relationships, it breaks down in the evaluation of outsourcing arrangements in which relationships require investments that can moderate the outcome. Organizations are not mere substitutes for structuring efficient transactions as they are established to govern certain kinds of economic activities through a logic that is very different from that of a market. TCT fails to recognize this difference, particularly when investments are made into developing organizational capabilities (Ghoshal and Moran, 1996). Additionally measures based on behavioral and relationship theories have also been developed (Sabherwal, 1999; Cross, 19

PAGE 31

1995; Huber, 1993). A measure of satisfaction was developed based on the Theory of Reasoned Action and used in a study by Heckmann and King (1994), where they identified positive and negative indicators of satisfaction using factor analysis. The major drawbacks of these studies is that outcome or success has been narrowly measured based on how the firm has defined the meaning of success, that is, were that firm's expectations realized. This has the problem of having made the construct "value laden". Moreover, these constructs of outsourcing outcome vary from organization to organization, with limited universal applicability. Hence it is difficult to utilize these measures in a more generic fashion across different outsourcing scenarios. Moreover the measurement constructs of outcome are not independent and hence can have adverse impacts on the measurement model during data analysis. For example costs reduction can be closely related to reduced service levels. Also it is seen that these measures are at the firm level and are potentially not suitable for a study at the functional or individual level of analysis. 20

PAGE 32

CHAPTER 3 THEORETICAL FOUNDATIONS As evident from the previous chapter, a growing concern among organizations, who are involved in outsourcing, is the post-deployment management of BPO or the "how" (Kern and Willcocks, 2000). The focus of this research project is into the "how" of post deployment BPO arrangements using the Resource Based View (RBV) of strategic management. RBV provides a research framework to select IS resources that are deployed in a domain of study and model their impacts on specific firm capabilities and performance outcomes. RBV has been successfully applied to multi-firm studies where the impact of inter organizational collaboration on value creation was studied (Lin, 2006). Several IS research papers discuss the relevance of studying the relational and infrastructure resources of BPO arrangements together (Koh, Ang and Straub, 2004). Information Technology is a facilitator of outsourcing through the concept of Net Enabled Organizations (Zahra and George, 2002). IT Infrastructure capabilities such as Voice over IP (VoiP) and networking, which makes cross-continental telecommunications inexpensive, data management systems like storage networks, distributed data databases and distributed application deployment/deployment platforms such as J2EE based servers like Weblogic, Websphere, etc. allow firms to pursue offshore outsourcing strategies very effectively. BPO is often coupled with the introduction of a large IT system such as an enterprise system like ERP or CAM (Broadbent, Weill and St. Clair, 2003). In fact the model of BPO is often termed synonymously in the literature as ITES or Information Technology Enabled Services. This concept of a technology driven outsourcing strategy is furthered by Sambamurthy, Bharadwaj and Grover (2003). who describe the phenomena as building dynamic capabilities and flexibility in the firm's 21

PAGE 33

processes and resources to allow the firm to react to changes in the environment. Clearly net enablement and IT infrastructure has created opportunities for the organization to exploit new strategies and react to new opportunities and tap into cheaper overseas labor markets. While studying the relevance of the inter-organizational infrastructure deployed in the BPO is apparent, there are also many justifications for studying the relationships among the client and vendor personnel. They include the following arguments -(1} the processes and systems do not operate in a vacuum, knowledge workers are important actors in the model and (2} there are two organizations involved in a BPO the client and the vendor, each with their own policies, their social systems and characteristics. Hence, several behavioral theories have been applied to the study of BPO relationships in studies at the organizational level, workgroup level or even at the individual employee level. Lyytinen, Mathiassen and Ropponen (1998} studied the risks involved in outsourcing and how the systems change the shape and behavior of the organizations and vice-versa. They found a significant reason for the failure of project implementations is due to the neglect of the relational goals of the organization as well as the social evolution of the organization, which is often overlooked in the outsourcing models. Social Capital Theory Social Capital Theory (SCT} has been proposed in many different terms in organizational research literature. Ghoshal and Moran (1996} describe social capital as the "organizational advantage". In summary, social capital theory is concerned with the significance of relationships as a resource for individual action. SCT is often contrasted with transaction cost theory that is based on human opportunism, while SCT is based on personal relationships developed through collective norms. 22

PAGE 34

Social capital comprises both the network and the assets that may be mobilized through that network. The social capital resource has three dimensions -(1) Group Identification, which is the structural or identification dimension, which refers to the network of interactions and how it builds group identity, (2) Generalized Trust or the relational dimension, which refers to the history of interactions among the people and how it influences their trust behavior and (3) Pro Sharing norms, which refers to those resources that provide shared meaning among parties. The RBV literature also recognizes the components of social capital resource as important IS resources (Feeney and Willcocks, 1998). They identify Pro sharing norms as the IS resource of Relationship building, which allows common language and conventions to build up so that diverse people can interact and work together. They state that "relationship building's most important contribution is in developing harmony of purpose and successful communications among staff', which resonates with the earlier definition of Pro sharing Norms. Another major component of IS resources is the trust that is developed between business units through a history of interactions, which can influence their knowledge sharing behavior (Nahapiet and Ghoshal, 1998). Recent research has shown that trust is an important resource for inter firm knowledge sharing (Jarvenpa and Staples, 2001). Social capital theory identifies Generalized Trust as an important relational dimension. Because the development of trust takes time, it can constitute a component of social capital resource and competitive advantage for the firm (Karimi, Somers and Bhattacharjee, 2007a). The final dimension of social capital resource to be considered in this study is Group Identification, which closely follows the definition of Partnership and Alignment resources described by Ross, Beath and Goodhue (1996). They describe the case of a 23

PAGE 35

large financial services firm that faced intense cost pressures, which they managed to overcome through strong business partnerships in many divisions. Such team alignments create group identification that allows a strong focus on the firm's priorities and outcomes. Thus, Group Identification is a strong IS relationship resource that needs to be considered in the domain of inter-organizational arrangements such as BPO. Social capital theory states that the social capital resource is owned jointly by the parties in a BPO relationship, and no one player has, or is capable of having, exclusive ownership rights. Moreover, although it has value in use, social capital resources cannot be traded easily. Social capital resources, therefore, are difficult to imitate because they develop over many years of cooperation sharing, identification and trust building. Further, relationship resources, such as social capital resources tend to be socially complex, relatively immobile, non-substitutable, inimitable, and rare (Wade and Hulland, 2004). Kankanhalli, Tan and Wei (2005) has applied SCT to model the contextual factors that help facilitate knowledge contributions to electronic knowledge repositories (EKR). They found that social capital theory can be applied to explain the behavior of resources embedded in inter-organizational environments such as networks of practice, where norms are formed to facilitate knowledge transfer. The benefits of applying social capital theory to a IS study stem from the ability to model and measure the creation of structural norms together with the development of relationship aspects such as generalized trust and group identification on other IS capabilities such as knowledge (Ghosh and Scott, 2007b). Of the other social and behavioral theories, the social exchange and organizational learning theories have been applied to the study of outsourcing as well (Kern and Willcocks, 2000). Social exchange theory (SET) is based on social psychology 24

PAGE 36

and sociology and the theory argues that a series of interactions generate obligations between people. Hence future action by a person is seen as being interdependent and contingent on the actions of another person. These interdependent transactions can also have the potential to generate high quality relationships under certain conditions (Cropanzano and Mitchell, 2005). IS researchers have utilized SET to study organizational knowledge transfer scenarios (Kankanhalli, Tan and Wei, 2005; Chen and Choi, 2005). The authors found that social and individual cost and benefit factors in knowledge management can be accounted for in SET. The application of SET is suitable when the participants and contributors of knowledge work under an assumption of a relatively longer term relationship of interest (Kankanhalli, Tan and Wei, 2005). Thus, a request for knowledge by one person, if met by the collaboration group at one point in time results in building an obligation for future reciprocation, supporting a link between creating social bonds and knowledge management in inter-organizational groups. Various organizational learning theories have been extended to study inter and intra organizational learning (Huber, 1991). The concept of organizational learning is also tied to organizational memory or the ability of the organization to remember norms and structures from its past experience whether good or bad. Organizational Memory is considered distinct from individual memory just as organizational learning is different from individual learning. These concepts are integral to organizational change, as these processes of learning and memory are what are used by the firm to meet challenges from the environment and internal forces and enhances its performance. The manifestations of these processes are often found in the decision making, innovation and information acquisition and distribution functions (Cohen and Leventhal, 1990; Andreu and Ciborra, 1999). 25

PAGE 37

The Task-technology fit literature also shows support for the link between both social capital resource and infrastructure resource on the sharing of knowledge and organizational performance (Becerra-Fernandez and Sabherwal, 2001). The theory of task-technology fit asserts that "for a deployed infrastructure to have a positive impact on individual performance the technology must be utilized and the technology must be a good fit with the tasks it supports." Their study shows that task inter-dependence and presenting opportunities for interaction through the infrastructure have a large role in knowledge combination and management. Therefore, this leads to the notion that if people's tasks, goals and objectives are more closely related, such that interactions are fostered, then it will lead to better combinations of the infrastructure resource and creation of relational norms and greater knowledge management will occur (Wasko and Faraj, 2000). Other studies have reported the benefits of relationship resources on inter firm cooperation and operations. Kumar, Dissel and Bielli (1998) pointed out the example of the textile merchants in Prato, Italy. They show how years of face-to-face operations and close business relationships have built up inherent social capital among a group of textile merchants in Prato, Italy and shaped their business practices over the years. This gradual evolution and "in-bred" relationships have allowed the textile industry in Prato, Italy to survive multiple shocks the period of fascism, two world wars, the shortage of raw materials and the emergence of cheap textile producing counties. Regardless of the environmental and market gyrations, the textile industry in Prato continues to thrive due to the great business and personal relationships and developed social norms, which has resulted in improved organizational performance among the merchants. 26

PAGE 38

The stream of research on virtual teams also supports the importance of relationship resources on inter organizational knowledge processes that span multiple locations. The BPO scenario typically puts non-collocated teams in two separate organizations -the client and the vendor to implement a business process. Research on virtual teams and the deployment and use of collaborative technologies have also investigated the inter-organizational phenomena of non-collocated teams building shared context and exchanging knowledge (Malhotra, et. al., 2001 ). This research supports the link between social capital resources and organizational performance. Moreover, these studies also investigate the inter-organizational dimension and raises the issue of an environment without common norms for knowledge management. Virtual team research has investigated the need to create a shared understanding among members before effective team practices can be established (Jarvenpaa and Leidner, 1999). Results indicate that such norms and knowledge management take place when planned opportunities for interactions are made between virtual team members. The development of a common language allowed better use of collaboration technologies in a case study of a virtual team at Boeing (Malhotra, et. al., 2001 ). The virtual environment also was found to facilitate more knowledge management than a regular collocated group when working norms are established. These findings are all consistent with knowledge management in the BPO scenario, where social and structural norms are formed. Their results indicate that such norms are established and knowledge management take place when planned opportunities for interactions are made between virtual team members. These studies note that the development of social capital requires the active and willing engagement of individuals within an infrastructure. The organization is better placed to create social capital through accessible infrastructure resource components. Wasko and 27

PAGE 39

Faraj (2005) have used SCT to evaluate participation and quality of knowledge contributions to an electronic network of practice. Their study found that infrastructure factors and organizational factors can impact the establishment of norms in the inter organizational relationships. IS researchers have also studied factors that improve BPO success rates and found that better management practices and selective outsourcing can improve outsourcing success (Lacity and Willcocks, 1998). Management practices are also seen to help foster relationships and improved outsourcing outcomes (Rottman and Lacity, 2004). Carmel and Agarwal (2002) identify several factors that can facilitate the alignment of teams across the client and the vendor. Support can also be drawn from the situated organizational learning literature which places emphasis on the context of leaning through the creation of structural norms (Lave and Wenger, 1991). This argument supports the relationship between establishing social capital and norms on knowledge management. Situated learning theory also claims that individuals can build shared context and understanding through simultaneous participation in group processes (Nidumolu, Subramani and Aldrich, 2001). They stress that using this perspective to study knowledge management provides emphasis on knowledge flows as opposed to merely knowledge stocks or the artifacts of knowledge management technologies. Researchers have also utilized situated learning to model and understand the situated nature of cognition, when structural norms are built as the network actors make sense of and act coherently in their world (Yuan and McKelvey, 2002). The above studies imply a clear link between the Social capital resources and Infrastructure resources on knowledge management capabilities and the combined impact of those two resources (termed as IS resources) and capabilities on BPO 28

PAGE 40

outcome. However, neither of these links have been tested in a BPO scenario in the existing IS research literature using the RBV framework. The theoretical foundations are summarized in Figure 3.1. Figure 3.1 Resource Based Social Capital Theory Social Exchauge Theay Vinual Teams Theoretical Foundations Of!anizational Learning Theory Coordinatioo Theory Virtual Teams KnO\\ tedge Management 29 Social Capital Theory Resource Based View lustitutional Theory Social Exchange Theory Management Models

PAGE 41

BPO Outcomes The Information systems research literature shows prior studies that have investigated outsourcing outcome using various theories Institutional theory (Teo, Wei and Benbasat, 2003), process theory (Carmel and Agarwal, 2002) and economic theories (Ang and Straub, 1998; Levina and Ross, 2003). Economic theory suggested that outsourcing relationships are successful in the cases where the client firm develops a complementary set of core competencies from the relationship and highlighted the need to focus on the information processing requirements for the organization, which needs to be matched with the level of uncertainty facing the organization in the BPO. This supports the need to measure BPO outcome using an information flow based measure. Carmel and Agarwal (2002) used process theory to understand the key elements for outsourcing success at each of the four stages of the client's offshore outsourcing strategy offshore bystander, offshore experimenter, proactive cost focus using outsourcing of non-core work and proactive strategic focus with outsourcing of core work processes. The authors reported the important role of the organizational profile in the strategic decision making process related to outsourcing by the client firm. Other studies have investigated the cost advantages that arise from the economies of scale and scope in the outsourced process possessed by the outsourcing vendor organization (Ang and Straub, 1998). Teo, Wei and Benbasat (2003) used institutional theory to understand the factors that enable the successful adoption of inter-organizational systems. They found that once the outsourcing arrangements are made, the coercive and normative pressures of cooperating between the client and the vendor firms will significantly influence the structure and cultural aspects of both firms. Therefore, we can clearly see that while the 30

PAGE 42

initial decision to outsource may be heavily dependent on cost-benefit analysis and to save operational costs, the outcome is significantly weighed towards the operational culture and working environments between the two firms -the client and vendor. Clearly economic theories can be used to make the outsourcing decision, however the success of the outsourcing decision often ends up in the ongoing process management of the BPO relationships which the economic theories tend to ignore (Ghosh and Scott, 2005a; Ghosh and Scott, 2006a, Ghosh and Scott, 2007b). However, strictly process based measures that do not account for the flow of knowledge/information in the outsourcing relationship, may not fully explain the observed outcome. A measurement using information flow can provide for a broader measure of the BPO outcome. Without the measurement of any vulnerability of the client firm from lack of adequate information flow, a true assessment of the BPO outcome is not possible (Ghosh and Scott, 2006a; Ghosh and Scott, 2005a; Ghosh and Scott, 2004). It is highly possible that client firms that place greater emphasis on establishing relationship capital reduce their vulnerability and may fall in the 46% that experience successful BPO outcomes (Ghosh and Scott, 2005a). Coordination Theory Coordination theory, which is based on the coordination, production and vulnerability costs associated with the interactions of organizational subunits (Malone and Crowston, 1999) has been used to study the impact of information technology on organizational structure (Malone and Crowston, 1999). A model based on coordination theory can shed light on the effective flow of information and knowledge and the allocation of resources, both of which are critical components of offshore BPO (Malone 31

PAGE 43

and Crowston, 1994). Coordination theory offers a vehicle to study the impact of technology on organizations through the medium of business processes-referred to in the theory as coordination processes. It is well suited to measure the costs of information/knowledge transfer arising in a BPO relationship (Ghosh and Scott, 2006a; Ghosh and Scott, 2004; Ghosh and Scott, 2005a). The advantage of coordination theory is that it allows the study of the organizational impacts of outsourcing at the micro-level"down in the trenches" using the coordination structures (Malone, 1987). A coordination structure is defined as a pattern of decision-making and communication among a set of actors who perform tasks in order to achieve goals. There are three kinds of costs for these coordination structures production costs, coordination costs, and vulnerability costs (Tushman and Nadler, 1978). Clearly, all three costs are impacted by outsourcing. The definition of each of the three costs are: 1. Production costs include the costs of running the outsourced business processes and will measure any efficiencies or deficiencies introduced. 2. Coordination costs are the costs to manage the communication between the client and the vendor firms. 3. Vulnerability costs are the unavoidable costs of a changed situation that are incurred before the organization can adopt to a new situation. These three costs form coordination theory form three dimensions of the theory and cumulatively reflect the overall process outcome (Malone, 1987). 32

PAGE 44

CHAPTER 4 DEFINING STUDY CONSTRUCTS Social Capital Resources The concept of social capital exists with the functional group and should be measured at the functional level as the creation and semantics of the construct are very much group or network related (Brewer, 2003). Social capital is multi-dimensional with each dimension contributing to the meaning of social capital where each dimension alone is not able to capture fully the concept in its entirety. The main dimensions are commonly seen as: (1) Trust or relational aspects together with types of social interactions, (2) Structural issues that relate to Identification with the group, and (3) Cognitive issues that include rules and norms governing social action (Coleman, 1988, Putnam, 1993). Nahapiet and Ghoshal (1998) also define three similar dimensions for social capital -(1) structural (referrals, timing, context, network ties), (2) cognitive (shared codes, language and narratives) and (3) relational (generalized trust, norms, obligations and identification). Researchers applying SCT have utilized these dimensions in operationalizing social capital in prior studies (Adler and Kwon, 2002, Kankanhalli, Tan and Wei, 2005). In IS literature, Kankanhalli, Tan and Wei (2005) have operationalized the social capital construct using three dimensions -(1) Generalized trust, (2) Identification and (3) ProSharing Norms. These dimensions match the three dimensions in Nahapiet and Ghoshal's (1998) definition of social capital (1) Relational, (2) Structural and (3) Cognitive. In the inter-organizational context, emphasis is placed on network configurations, shared cognitive norms and trust (lnkpen and Tsang, 2005). BPO closely resembles network alliances that involve inter-member social ties as a foundation for exchange and sharing of knowledge (Gulati, Nohria and Zaheer, 2000). Hence the 33

PAGE 45

social capital resource construct must be measured to capture the concept as a reflection of three interacting dimensions -(1) Group Identification, which is the structural or identification dimension, which refers to the network of interactions and how it builds group identification, (2) Generalized Trust or the relational dimension, which refers to the history of interactions among the people and how it influences their trust behavior and (3) Pro Sharing norms, which refers to those resources that provide shared meaning among parties (Nahapiet and Ghoshal, 1998). Group Identification Group Identification is a condition where the interests of the individuals merge with the interests of the organization, resulting in the creation of an identity based on those interests. Identification sets the context with which communication and knowledge exchange occur among organizational members (Nahapiet and Ghoshal, 1998). Three components of identification that have been used in IS research are -(1) sharing of common values and (2) common goals and (3) strong membership towards the group (Kankanhalli, Tan and Wei, 2005). When Identification is strong, the effort required for knowledge contributors may not be a deterrent to knowledge contributions and would therefore be a facilitator for knowledge management capabilities in a BPO arrangement. Identification is an extremely important component of the social capital resource in offshore BPO arrangements due to the natural remoteness and lack of the usual organizational camaraderie in such scenarios. Generalized Trust Generalized trust is an impersonal form of trust that does not rest with a specific individual but rather is a property of the behavior that is generalized to a social unit as a whole, such as a group of knowledge workers in a BPO arrangement (Putnam, 1993). 34

PAGE 46

When generalized trust is strong, the effort required for knowledge sharing may not be salient to knowledge contributors because they believe that knowledge shared is not likely to be misused by the other party (Davenport and Prusak, 1998). Conversely when generalized trust is weak, knowledge contributors may find the effort required for knowledge sharing to be high as they believe that others may inappropriately use their knowledge. Generalized trust has been viewed as a key factor that provides a context for cooperation among multiple units (Nahapiet and Ghoshal, 1998). The generalized trust resource has three components that help define its formation and existence-(1) the establishment and keeping of obligations, (2) giving credit to people for contributions where credit is due, and (3) using other worker's or unit's knowledge appropriately (Kankanhalli, Tan and Wei, 2005). Generalized trust is an extremely important IS resource in the offshore BPO context for effective knowledge exchange because of the absence of adequate personal knowledge among the collaborators and coworkers in the client and vendor firms. Pro Sharing Norms A norm represents a degree of consensus in the social system (Coleman, 1990). Norms have the effect of moderating human behavior in accordance with the expectations of the group or community. Pro sharing norms therefore are likely to enhance the climate for knowledge sharing and positively impact knowledge management capabilities in the firm. Components of the pro sharing norms resource include (1) teamwork collaboration and sharing (Jarvenpaa and Staples, 2000); (2) willingness and ability to value and respond to diverse views and opinions among staff and (3) the tolerance for mistakes and failure in the relationships. When Pro sharing norms are strong, the costs of knowledge sharing may not be a deterrent to knowledge 35

PAGE 47

contributors (Kankanhalli, Tan and Wei, 2005). Pro sharing norm enable the workers from diverse backgrounds and remote firms client and vendor in the BPO to forge common language, conventions and values and leads to improvements in knowledge management capabilities in the BPO arrangement. Hence Pro Sharing Norms are an important component of the social capital resource in the context of offshore BPO. Infrastructure Resources Infrastructure resources refer to a firm's shared IT assets (e.g., hardware, software tools, and networks, databases and data centers). They are the foundations for a firm's IT architecture, which is the plan or design that supports multiple service processes, products and groups of users groups across the firms. From a strategic resource perspective for building strategic level capabilities, the individual IT asset is commodity-like, widely available, imitable, and relatively easy to obtain, and is therefore, incapable of generating long-term economic rents, and generally not found to be a source of sustained competitive advantage (Wade and Hulland, 2005). Reliable IT infrastructure resources, however, can ensure the success of IT architecture, which tends to be highly firm specific and evolves over a long period of time during which gradual enhancements are made to reflect changing business needs (Karimi, Somers and Bhattacharjee, 2007a). Knowledge Management Capabilities Drawing on the research stream in knowledge management capabilities, knowledge management is defined through the five processing steps involved: (1) Acquisition, (2) Application, (3) Conversion (4) Transfer and (5) Protection (Szulanski, 1996; Gold, Malhotra and Segars, 2001; Tanriverdi, 2005). The actions that result form the above consist of one or both parties seeking knowledge and/or providing knowledge 36

PAGE 48

in response to the request, such that one or both parties are affected by the experience (Huang and DeSanctis, 2005, Szulanski, 1996). The facets of knowledge sharing in a BPO are: (1) one or both parties seeking to acquire knowledge, (2) one or both parties conver:ting tacit knowledge or pointing to the location of already explicit knowledge in response to the request, (3) one or both parties transferring the knowledge and (4) the seeking party applying the new knowledge. This knowledge can be related to either the outsourced process and/or product/market(s) that are being served by the process. Finally, knowledge must also (5) be protected against unauthorized access and use on both sides of the outsourcing arrangement. BPO Outcomes Coordination theory offers a vehicle to measure organizational performance through business processes referred to in the theory as coordination structures (Malone, 1987). These information flows and coordination structures manifest in three types of information costs. Hence to measure a process outcome, the measure must capture the reflection of the information flow of the coordination structure of the process on the three information costs Production, Coordination and Vulnerability (Ghosh and Scott, 2006a). There are three types of information costs associated with each coordination structure: 1. Production costs are the "transaction" costs of running the outsourced processes and they measures any efficiencies or deficiencies introduced. To evaluate BPO outcome, the change in the client's transaction costs after the BPO implementation must be measured. 2. Coordination costs are the "management" costs to manage the communication between the client and the vendor firms, prioritization of 37

PAGE 49

activities and the allocation of resources. To evaluate BPO outcome, the change in the client's coordination costs after the BPO implementation must be measured. 3. Vulnerability costs are the "strategic" costs associated with a delayed response to a changed market situation. The client firm in a BPO is further removed from the markeVend user resulting in missed market knowledge. This leads to additional vulnerability costs for the client. To evaluate BPO outcome, the change in the client's vulnerability costs after the BPO implementation must be measured. The combination of the three information costs derived from coordination theory is termed as BPO outcome (BPOC). The survey instruments are listed in Table 4.1. 38

PAGE 50

Table 4.1 Survey Instruments SOCIAL CAPITAL RESOURCE (SCR) SCR 1: There is a norm of teamwork among staff in the BPO relationship. SCR2: There is a norm of openness of diverse/conflicting views among staff in the BPO relationship SCR3: There is a norm of tolerance of mistakes among staff in the BPO relationship SCR4: Client and vendor staff in the BPO relationship share common values. SCR5: Client and vendor staff in the BPO relationship share common goals SCR6: Client and vendor staff in the BPO relationship have strong group identification SCR7: Client and Vendor staff in the BPO relationship give credit to people where it is due SCAB: Obligations between organizations are always sustained by staff in the BPO relationship SCR9: Client and Vendor staff in the BPO relationship use other's knowledge appropriately INFRASTRUCTURE RESOURCES (IR) IR1: Appropriate hardware, software, and network infrastructures are in place to support the BPO IR2: Appropriate knowledge systems and tools are in place in the infrastructure to support the BPO IR3: Access technologies are administered in the infrastructure to allow appropriate access to knowledge (*Item dropped) KNOWLEDGE PROCESS CAP ABILITIES Acquisition ACQ1: Our organizations have processes for acquiring knowledge from our BPO partners ACQ2: Our organizations have processes for generating new knowledge from existing knowledge in the BPO relationship ACQ3:0ur organizations have processes for exchanging knowledge between individuals in the BPO relationship. KNOWLEDGE MANAGEMENT CAPABILITIESConversion CNV1: Our organizations have processes for converting knowledge into the design of future BPO support services CNV2:0ur organizations have processes for distributing knowledge throughout the two organizations in the BPO relationship CNV3: Our organizations have processes for integrating different sources and types of knowledge (*Item dropped) 39

PAGE 51

Table 4.1 Survey Instruments (cont.) KNOWLEDGE MANAGEMENT CAPABILITIESApplication APL 1 : Our organizations have processes for applying knowledge learned from experience APL2: Our organizations have processes to make knowledge accessible to those who need it APL3: Our organizations have processes to quickly link sources of knowledge in solving problems KNOWLEDGE MANAGEMENT CAPABILITIESTransfer TRF1: I am encouraged to discuss my work with employees in other workgroups TRF2: Benefits of sharing knowledge in my organization outweigh the costs TRF3: Knowledge is freely exchanged freely through email, presentations, newsletters, etc. KNOWLEDGE MANAGEMENT CAPABILITIESProtection PRT1: Our organizations have processes to protect knowledge from inappropriate access in the BPO relationship PRT2: Our organizations have processes to protect knowledge from inappropriate use in the BPO relationship PRT3: Our organizations clearly communicate the importance of protecting knowledge BPO OUTCOMEVulnerability VUL 1: After the BPO, my organization adapts more quickly to unanticipated changes VUL2: After the BPO, my organization can react more quickly to new market info VUL3: After the BPO, my organization's response to market changes involves alesslengthyprocess BPO OUTCOME-Production PRD1: After the BPO, the transaction costs related to these outsourced processes are lower PRD2: After the BPO, we can serve our customers more effectively now PRD3: After the BPO, less resources are required to operate the outsourced processes BPO OUTCOMECoordination CRD1: After the BPO, a decreased number of resources are needed to support the outsourced process CRD2: After the BPO, a decreased number of managerial contacts are needed to support the outsourced process CRD3: After the BPO, less time is required to get status and updates about the outsourced process 40

PAGE 52

CHAPTER 5 RESEARCH HYPOTHESES The following subsections build and present the research hypotheses of this study. A research model is developed to study the impact of social capital and infrastructure resource on the knowledge management capabilities in the BPO relationship and the BPO outcome. The research model is illustrated in Figure 5.1. Figure 5.1 Research Model 41

PAGE 53

As reported earlier, Wasko and Faraj (2005) found that motivational factors and organizational enablement impact the knowledge management in the relationship. Kumar, Dissel and Bielli (1998) described the textile merchants in Prato, Italy. Their study showed how face-to-face operations and closeness of business relationships have built enablers for knowledge management in their business practices over the years. Such close interactions and effective knowledge management can be established in a non-collocated team as well, when social capital resources are established. Malhotra, et. al. (2001) reported that the development of common language and norms between non collocated teams allows better use of collaboration technologies and knowledge management in a case study of a virtual team at Boeing. Such a virtual environment also was found to facilitate more knowledge management than a regular collocated group. In their study of knowledge transfer from consultants to clients, Ko, Kirsch and King (2005) found the impact of several components on the effective management of knowledge in inter firm relationships. They identified factors such as communication encoding and decoding capacity that builds a shared understanding that impacts knowledge transfer. Szulanski (2000) identifies the importance of "bridging the communication gap, the coding schemes and cultural conventions" as critical to overcoming knowledge management stickiness. This study also claims that knowledge transfer occurs as social capital is built up and documented. Other studies have investigated virtual communities or remote networks of practice (Brown and Duguid, 2001) and found that the social capital resource in the relationship significantly impacted the quality of knowledge contributions. Social capital builds trust and commitment and is a facilitator for collective action when multiple parties are engaged. 42

PAGE 54

Knowledge management is more likely when social relationships and generalized trust are strong (Szulanski, 1996; Ko, Kirsch and King, 2005). Social capital resources allow strategic networks to be built that facilitate knowledge management (Gulati, Nohria and Zaheer, 2000). These networks are seen to encourage group identification. lnkpen and Tsang (2005) also found that behavioral factors play a significant role in knowledge transfer in these social networks. The stream of research in partnerships and inter organizational learning has also identified that social capital resources in the relationship play a role in knowledge transfer in both formal and informal settings (Malhotra, Gossain and El Sawy, 2005; Argote, 1999). Such collective action encourages the development and management of knowledge as group identification is built up. The above arguments support the first hypothesis (H1), which is stated as follows. Note that H, is shown in Figure 5.1 as a link from Social Capital Resources (SCR) to Information Systems Resources (ISR), however, the hypothesized relationship that is tested is a link from SCR to Knowledge Management Capabilities (KMC). Please refer to the direct effects model in Figure 8.1. H1 Social capital resources have a positive association with knowledge management capabilities in the BPO organization. IT infrastructure resources have been linked to a firm's ability and willingness to develop innovative business applications and knowledge management capabilities (Davenport and Prusak, 1998). Such applications can be in quite far reaching areas such as business intelligence for knowledge discovery, case based reasoning for knowledge application, content management and servers for knowledge codification, etc. They support and fully facilitate the firms knowledge sharing capabilities (Karimi, Somers and Bhattacharjee, 2007a). IT infrastructure resources are critical in developing, building, and 43

PAGE 55

assimilating IT capabilities, and in enhancing IT's productive value by aiding implementation, simplifying system integration across diverse applications, and creating economies of scale and scope in system maintenance. In contrast, lack of IT infrastructure resources severely restricts a firm's IT capabilities and increases the costs of building or supporting IT (Ray, Muhanna and Barney, 2005). Clearly adequate infrastructure resources (e.g., hardware, software, network, and server and database technologies) need to be planned, put in place, and reliably available well in advance both for pre-and-post BPO implementation stages. The above arguments support the second hypothesis (H2), which is stated in the alternate form as below. Note that H2 is shown in Figure 5.1 as a link from Infrastructure Resources (IR) to Information Systems Resources (ISR), however, the hypothesized relationship that is tested is a link from IR to Knowledge Management Capabilities (KMC). Please refer to the direct effects model in Figure 8.1. H2 Infrastructure resources have a positive association with the knowledge management capabilities in the BPO organization. Resources are essential raw materials for capability-building, and their availability determines a firm's ability to build such capabilities, which are often critical drivers of firm performance. Resource synergies and the significance of integration, combination, and co-specialization of resources have been emphasized in past research (Tanriverdi and Venkataraman, 2005). Further, some capabilities may derive from a contribution of a single resource, while others may require highly complex interactions involving the cooperation of many different resources (Ray, Muhanna and Barney, 2005). A review of the prior research on BPO shows support for the combined IS resources to impact the knowledge management capabilities. Levina and Ross (2003) 44

PAGE 56

have used economic theory to suggest that outsourcing relationships are successful in the cases where the client firm maintains proper organizational infrastructures to manage the relationship between the client and vendor and develop complementarities. This study has also reported that the information processing requirements for the client organization need to be matched with the level of uncertainty facing the organization in the BPO. Lacity and Willcocks (1998) studied the allocation of resources and found that infrastructure resources must be properly distributed to meet organizational requirements and emphasize contributions to knowledge repositories. Outsourcing relationships are usually set up to have a loose relationship between the firms. Such a loose alliance creates a fragmented internal environment, which can hamper the knowledge flow and cause the BPO client to become vulnerable to missing key knowledge from the vendor side similar to loosely coupled environments (Orton and Weick, 1990), which require IS resources to support the knowledge management system in the BPO. When a BPO is pursued, additional infrastructure needs to be established so that the organization can effectively exchange knowledge within the BPO context (Levina and Ross, 2003). Hence the effectiveness of the knowledge management capabilities of the firm after the BPO deployment will depend on the effective pairing of other relationship based resources with IT infrastructure resources that is accessible to both sides of the BPO client and vendor. Similarly, because of resource synergies, individual dimensions of IS resources should not be viewed in isolation. As a collective, mutually reinforcing, and a higher order factor structure, which accounts for the relationships among social capital resources and IT infrastructure resources, IS resources are expected to have a combined impact on building knowledge management capabilities. The above arguments support the third hypothesis (H3), which is stated as: 45

PAGE 57

H3 Information Systems (IS) Resources have a positive association with knowledge management capabilities in the BPO organization. BPO vendors allocate their resources over multiple clients. Thus, BPO environments are unique in that client initiated knowledge management capabilities will not only benefit the initiating client and its vendor, but other clients of that vendor as well. However, knowledge created through the mobilization of social capital resource is retained in the work group (Nahapiet and Ghoshal, 1998). Such assets are jointly owned and symmetrically transferred within the BPO work team established by the client and vendor. Hence, positive benefits can be measured in the client organization (Ghosh and Scott, 2006a). It is therefore advisable to measure the impact of knowledge management capabilities on the process cost structure of the client. Knowledge management capabilities reduce the efforts involved with coordination of BPO activities (Ghosh and Scott, 2006a). Productions costs related to the operation of the process that is outsourced are also lowered with the increase in knowledge transfer since improved performance is to be expected from the vendor as they become more knowledgeable. These benefits arise as the vendor can access knowledge and reuse it more rapidly (Watson and Hewett, 2006), hence lower production costs result. Creating opportunities for knowledge management results in more informal exchanges of information among personnel in the two firms. This effectively improves coordination capabilities (Ghoshal, Korine and Szulanski, 1994). The research stream in project management also identifies the flow of knowledge as a critical resource in coordination activities (Snider and Nissen, 2003). The greater the scope and frequency of knowledge management, the smoother the project coordination tasks such as resource 46

PAGE 58

allocation and task prioritization. In a BPO, coordination costs will be lowered as knowledge management increases. Outsourcing arrangements are usually set up to have a loose relationship between the firms. Such a loose alliance creates a fragmented internal environment, which can hamper the partnership culture and information flow and cause the BPO client to be vulnerable to market changes (Lacity and Willcocks, 1998). In a typical BPO the client often relinquishes some of the customer interactions that are now managed by the vendor. Mechanisms need to be put in place to facilitate the management of customer knowledge that is collected by the vendor in their customer interactions back to the client. Therefore, with an increase in knowledge management, it is anticipated that vulnerability costs will be also be lowered (Ghosh and Scott, 2006a). Therefore the impact of increasing the knowledge management in the BPO relationship is to reduce all three information costs which improves the BPO outcome. The above arguments support the fourth hypothesis (H4), which is stated as: H4 Knowledge management capabilities have a positive association with BPO outcomes in the BPO organization. Capabilities are special types of resources that are non transferable between firms and whose purpose is to improve the productivity of other resources. Resources and capabilities complement each other and firms need to effectively select and match the right resources to build the needed capabilities. Moreover to support building those capabilities, the organization needs to effectively deploy those resources as well (Melville, Kraemer and Gurbaxani, 2004). In addition, certain organizational capabilities may be complementary to the IS resources in generating IT business value. Melville, Kraemer and Gurbaxani (2004) examined the nature of resource synergies and further 47

PAGE 59

suggested that human IT resources are complementary to IT infrastructure resources and they both create economic value for a foe a firm by conferring operational efficiencies that vary in magnitude and type depending upon the organizational and technological context. Their study further suggested that (1) there are complementary relationships between certain IS resources in generation of IT business value, and (2) the existence and magnitude of the complementarity between any two specific instantiations of these resources varies depending upon the organizational and technological contexts. IT infrastructure and e-commerce capability, for example, have been shown to have positive complementarity effects on some measures of firm performance (Zhu, 2004). These studies suggest that IT infrastructure resources can provide the platform to deploy knowledge management capabilities in the BPO. Further the effectiveness of those knowledge management capabilities as seen on the BPO outcome may be enhanced by the amount and the combination of IS resources of social capital and infrastructure that are deployed. The social capital resource affects the BPO outcome by impacting the information costs of production, coordination and vulnerability through the knowledge management capabilities. Social capital can reduce the cost of an item of work in the outsourced process by economizing on information and coordination effort (Nahapiet and Ghoshal, 1998). A study by Koh, Ang and Straub (2004) has stressed the need to have psychological contracts, i.e., social capital within individuals that can drive the behavior of vendor personnel, such as such as sharing knowledge among individuals in the BPO arrangements. The same arguments have been found for the Infrastructure resource, which enable knowledge management capabilities by allowing the storage, search, 48

PAGE 60

administration of access rights and retrieval of knowledge resources both within and across organizations (Scott, 2000). Infrastructure supports building inter organizational capabilities, however the strength of the capability and its impact on the performance of the inter-firm BPO arrangement depends on the relationship resources deployed (Scott, 1998). A simple IT system such as a transaction system is enough to capture information but is inadequate for knowledge management, which needs human resources in combination with those infrastructure resources to build, assimilate and leverage those knowledge entries into the business processes (Scott, 1998). Hislop (2002) points out that technology infrastructure resources can facilitate capabilities for explicit knowledge, while relationship resources are needed for tacit knowledge capabilities. Clearly a complementary relationship exists among the IS resourcessocial capital resources and infrastructure resources and KM capabilities on the BPO outcome. Any weakness in these IS resources will undermine the value derived from building those knowledge management capabilities in the BPO. These arguments suggest that there is a strong theoretical foundation to support a complementary relationship among IS resources and knowledge management capabilities that impacts BPO outcome. Any weakness in IS resources -social capital and infrastructure can reduce the benefits produced by knowledge management capabilities on BPO outcomes. Likewise higher IS resources will promote more effective knowledge management and will lead to better BPO outcomes. Therefore, the fifth hypothesis (H5 ) can be stated as: H5 -Greater IS Resources in conjunction with higher knowledge management capabilities are associated with improved BPO outcomes in the BPO organization. 49

PAGE 61

CHAPTER 6 RESEARCH METHODOLOGY Research Design The methodology used is a field study. The study was conducted in two stages. The first stage involved a pilot study to evaluate the survey-the wording and descriptive explanations on the survey. The second stage involved a survey to collect data from client side employees. Knowledge workers in multiple supervisory groups involved in the operation of customer service and network consulting processes within the client organization in a high technology outsourcing environment were surveyed. The survey measured the actual "presence" of IS resources and the knowledge management capabilities that take place between the two organizations in the BPO arrangement. The survey also measured the outcomes of outsourcing reflected in the three types of costs: vulnerability, coordination and production costs. The quantitative data that was collected from the surveys was analyzed with PLS-Graph software tool using the Partial Least Squares method of structural equation modeling (Chin, 1998). Study Variables The independent variables for this study are the Social Capital Resources (SCR) and the Infrastructure Resources (lA), which form a formative construct termed IS Resources (ISR). ISR is a second order formative construct, which is formed by the above two first order constructssocial capital resources and infrastructure resources. The dependent variable is the BPO outcomes (BPO) as measured by the three coordination costs production, coordination and vulnerability. Knowledge management capabilities (KMC) is the additional variable measured in the research model. The data 50

PAGE 62

was collected at the worker level in the client organization. Knowledge workers rated their perceptions of the variables. Items were scored on a 5 point Liekert scale of Strongly Agree (5). Agree (4), Neutral (3), Disagree (2) and Strongly Disagree (1 ). Measurement of Research Variables All research instruments used in the study were adopted from prior research studies. Hence they were pre-validated research instruments that only needed content validation for their use in the BPO domain. The Knowledge management capabilities (KMC) construct is a second order factor that is measured with five reflective first order constructs with 3 indicators each. The five constructs are: knowledge acquisition (ACQ 1 3), knowledge conversion (CNV1 3), knowledge application (APL1 3), knowledge protection (PRT1 3 ) and knowledge transfer (TRF1 3). The individual items were adopted from the knowledge management capabilities model developed and validated by Tanriverdi (2005) and Gold, Malhotra and Segars (2001). The components of knowledge management capabilities in a BPO are: (1) one or both parties seeking to acquire knowledge, (2) one or both parties converting tacit knowledge or pointing to the location of already explicit knowledge in response to the request, (3) one or both parties transferring the knowledge and (4) the seeking party applying the new knowledge. This knowledge can be related to either the outsourced process and/or producVmarket that is being served by the process. Finally, knowledge must also (5) be protected against unauthorized access and use on both sides of the outsourcing arrangement. IS resources (ISR) is a formative index formed by two reflective first order constructs social capital resources (SCR) and IT infrastructure resources (IR). Social capital resources (SCR) were measured by nine reflective items, three each from each of the three dimensions pro sharing norms, group identification and generalized trust. The 51

PAGE 63

definition of the social capital construct by Nahapiet and Ghoshal (1998) specifically states that the these dimensions cannot be easily separated and the that the social capital manifests into these three dimensions Three items reflect the Pro sharing norms dimension (SCR1 3 ) which tapped into the worker's acceptance and use of cognitive norms in knowledge sharing in the BPO. The group identification dimension was reflected using three items (SCR4 6 ) that tapped into the strength of group unity and common goals among the staff on the BPO. The generalized trust dimension of SCR was reflected using three items (SCR7 9 ) that probed into the relational aspects, which refers to the history of interactions among the people and how it influences their trust behavior. These instruments were adopted from the social capital study on electronic knowledge repository usage developed and validated by Kankanhalli, Tan and Wei (2005). IT infrastructure resources (IR) were measured using three items (IR1 3 ) that assessed whether hardware, software, network, and the necessary server and database technologies, knowledge tools and access restrictions were in position prior to the BPO deployment 52

PAGE 64

CHAPTER 7 DATA COLLECTION Collection Methodology The client organization chosen is a multinational technical support organization. The organization is a leader in the support of multi-vendor networking equipment, with network management and service capabilities. Network service offers include design, installation, monitoring and break-fix support. The product introductions, field trials, network design and installation services were outsourced to an offshore vendor to increase available headcount for these processes, reduce operating costs through labor arbitrage and provide in-region internationally located technical personnel. Organizational Practices for Knowledge Management The chosen organization was found to have several KM systems and practices to provide bi-directional transfer of knowledge from the client to the vendor and vice versa. In this BPO relationship, the client personnel are more experienced in the work domain than the vendor personnel. The client organization realized that to make the BPO effective vendor personnel needed to be mentored and trained in the domain. Hence, the heavier knowledge transfer involved the training and development of vendor personnel to prepare them for the technical work involved in the BPO. The lesser knowledge transfer occurred from the vendor to the client in capturing customer site-specific information, as the vendor is closer to the installation locations. The mentoring resources on the client side were limited and needed to be managed effectively. To serve the two fold goals of providing training to the vendor personnel as well as supporting the bi-directional knowledge transfer, the client decided to institute practices to build social capital. A selection program was instituted, whereby 53

PAGE 65

vendors could apply for client mentoring and assistance on an upcoming project. The client established a review board to screen each request for knowledge potential and optimal fit for the goals. If a project is selected, then personnel from the vendor and client work collaboratively on the project over a span of 4-6 weeks. Advantages are seen in building one or more social capital resources-in terms of exchange of norms, work practices, common language and standards as well as tacit knowledge that is difficult to codify and contextualize. Relational resources described earlier are built up as obligations are set and met, resulting in the development of trust, understanding of cultural diversity and establishment of common identification and joint ownership for work. The latter was critical, as the BPO goals were broader than cost savings and also involved building competencies. Project meetings are held frequently and the client personnel served as mentors in the relationship thus there was an expectation of value to be obtained from the process on the part of the participating vendor staff. Likewise, the client participant was able to build up site specific and international expertise. Both parties, client and vendor, are motivated to participate in the program to improve network service quality and minimize down stream customer network issues. 54

PAGE 66

Outsourced Processes Survey respondents fell into one of two services processes customer service or networking consulting. Both processes shared the IT infrastructure, knowledge management capabilities and worked with the same vendor firm on the offshore side of the BPO relationship. The flow of these two processes are given below. Customer Service Process Customer Service consists of fielding customer reported troubles, collecting information and investigating and resolving those customer issues. This process currently has domestic staff as well as staff at the vendor site offshore. The client staff has the added responsibility of training and mentoring the vendor staff so they come up to speed on the process and work. The primary reasons for pursuing outsourcing for this process include: (1) to provide local staff for supporting regional customers in their time zone, (2) creating a 24 hour support day by using global resources, (3) reducing the operating expenses of running the process by utilizing cheaper offshore staff, (4) capturing and institutionalizing localized knowledge from vendor staff and (5) increasing the headcount of trained personnel on the process. The steps of the process are: 1. Receive customer problem notification either in the form of a customer call, an email, web based problem report. 2. The knowledge worker then makes contact with the customer to collect additional information about the problem, the products involvement, the customer's environment, etc. 3. The worker then examines the problem, including various research activities such as looking through product manuals, setting up and working on lab 55

PAGE 67

resources, consulting various knowledge bases, listservers, FAQ, etc and discussing with experts. 4. The worker can repeat steps 2 and 3 multiple times as more information is needed or as details are collected. This is very much adhoc processing, where very detailed and sensitive information about the customer is collected. The steps to be followed can be very different based on the information collected. 5. The worker has to take extreme precaution to not disrupt the customer's network too much and as well as not to hamper working sections of the network or products that are in service. The information is quite sensitive as they represent configuration data from working product installations. 6. Once a solution has been identified, the worker then performs the necessary steps to solve the customer's problem. Network Consulting Process Network Consulting consists of fielding customer requests for information and delivery of proposals and sales configurations in support of the customer request. The process involves collecting information at a high level and understanding the customer's needs to develop a proposal to be presented to the customer. This process currently has domestic staff as well as staff at the vendor site offshore. The client staff has the added responsibility of training and mentoring the vendor staff so they come up to speed on the process and work. The primary reasons for pursuing outsourcing for this process include: (1) to provide local staff for supporting regional customers in their time zone, (2) reducing the operating expenses of running the process by utilizing cheaper offshore staff, (3) capturing and institutionalizing localized knowledge from vendor staff and (4) increasing the headcount of trained personnel on the process. 56

PAGE 68

The steps of the process are: 1. Receive customer request for information and product/solution suitability either in the form of a customer call, an email, web based request. 2. The knowledge worker then makes contact with the customer to collect additional information about the request, their environment, etc. 3. The worker then examines the request using various configuration and design templates, product manuals, consulting various knowledge bases and discussing with experts. 4. The information collected by the worker is generally high level and often a "best guess" or approximate and hence is not as sensitive to the customer that would require extra security measures. 5. The worker generates a well structured proposal in a standard format to present the information and solution options to the customer. Survey Results The final survey consisted of 4 demographic questions and 36 items. The survey was performed using an online survey. Email solicitations were sent to 200 client knowledge workers with a link to the survey, which was active for one week. A convenience sample was drawn from a list of professional contacts. The 200 employees fell into 14 supervisory groups. Each group was responsible for supporting one or more products or product families or market segment. A total of 119 employees responded with completed surveys for a response rate of 60%. The final set of 119 respondents were not identified and therefore cannot be traced to a supervisory group. The survey requested that they identify which one of the two processes that they followed in their work customer service or network consulting. 57

PAGE 69

The respondents fell into one of two processes customer service (72 respondents) and network consulting (47 respondents). Comparisons of demographic factors of years of experience, years on the current job, years of education and gender was made between the respondents on the two processes and no significant difference were found using T-Tests (Table 7.1). The descriptive statistics for the data is shown in Table 7.2. The mean statistic, standard error and range and the skewness and kurtosis measurements suggest that the measurements do approximate normal distributions. 58

PAGE 70

Table 7.1 Demographics across Processes Characteristic Customer Network Consulting T-Value Service [72) (47) Mean Std Mean Std Years on Job 5.40 2.84 5.51 3.66 .18 (0.857) Professional 9.01 6.69 8.35 6.95 -.52 (0.604) Experience Years of Schooling 15.17 2.55 15.09 2.84 -.16 (0.871) Gender M(48), F(24) M (35), F(12) X"= 1.18; P=0.277 59

PAGE 71

0) 0 PSNt PSN2 PSN3 t01 t02 103 GT1 GT2 GT3 IR1 IR2 IR3 ACQ1 A C02 ACQ3 CNV1 CNV2 CNV3 APL1 APL2 APL3 PRT1 PRT2 PRT3 TRF1 TRF2 TRF3 VULt VUL2 VUL3 PROt PR02 PR03 CR01 CR02 CR03 Valid N (listwise) N Statistic 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 119 Ra11!1e Minimum Maximum Statistic Statistic Statistic 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 3 2 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 4 5 3 2 5 4 5 4 5 Descriptive Statistics Mean Std. Variance Statistic Std. Error Statistic Statistic 2.80 .097 1.054 1.112 3.24 .093 1.014 1.029 2.82 .078 .850 .723 2.99 .097 1.054 1.110 3.29 .081 .886 .786 2.68 .086 .938 .880 3.22 .072 .783 .613 3.29 .083 .903 .816 3.35 .076 .829 .688 3.38 .107 1.164 1.356 3.34 .100 1.091 1.191 3.86 .072 .784 .615 3.08 .090 .984 .969 2.79 .086 .938 .879 3.47 .077 .842 .709 3.32 .082 .892 .795 3.39 .095 1.035 1.071 2.94 .094 1.028 1.056 3.02 .078 .854 .729 3.18 .070 .766 .587 2.94 .085 .932 .869 3.60 .086 .942 .887 3.22 .077 .835 .698 3.55 .085 .927 .859 3.18 .087 .945 .892 3.25 .096 1.051 1.105 3.01 .097 1.054 1.110 2.85 .108 1.183 1.401 2.63 .096 1.049 1.099 2.76 .101 1.102 1.215 3.34 .085 .925 .855 2.99 .103 1.124 1.263 2.90 .103 1.123 1.261 3.50 .Q78 .852 .727 3.33 .092 1.001 1.002 3.59 .087 .951 .905 Skewness Statistic Std. Error .414 .222 .105 .222 .432 .222 .459 .222 -.320 .222 .557 .222 -.838 .222 -.389 .222 -.743 .222 -.617 .222 -.585 .222 -.278 .222 -.425 .222 .057 .222 -1.162 .222 -.386 .222 -.250 .222 .310 .222 .051 .222 -.083 .222 .055 .222 -.594 .222 .101 .222 -.682 .222 .006 .222 .100 .222 .248 .222 .267 .222 .343 .222 .211 .222 -.416 .222 -.020 .222 .348 .222 -.514 .222 -.390 .222 -.768 .222 Kurtosis Statistic Std. Error -.788 .440 -.826 .440 -.438 .440 -.900 .440 -.819 .440 -.599 .440 .953 .440 -.023 .440 -.687 .440 -.613 .440 -.540 .440 -.315 .440 -.862 .440 -.149 .440 .552 .440 -.824 .440 -.542 .440 -.873 .440 -1.035 .440 -.208 .440 -.883 .440 -.109 .440 -.253 .440 -.090 .440 -.740 .440 -.785 .440 -.576 .440 -.913 .440 -.708 .440 -.899 .440 -.613 .440 -.836 .440 -.812 .440 -.578 .440 -.368 .440 .148 .440 -t I 0" ii' ...... N 0 CD (/) (") .., ": en or (jj" c:r (/) 0 -3 (/)

PAGE 72

Knowledge Resources Data was collected on the usage of the following knowledge management systems over an 8 month period, while the knowledge management system was operational. (1) Listservers, Discussion Boards, FAQ that captured threads of discussions on topics raised by team members and their subsequent contributions. On average 29 unique issues or threads of discussion have been activated per month over the period. (2) Checklists -to guide technical support work from past experience and ensure that adequate data collection and situational analysis is being done. A total of 33 checklists have evolved in the same period ranging from building best practices in directed tasks such as configuring phones to open-ended tasks such as collecting customer information for network designs. (3) Lessons Learnt Lists -to ensure that new "rules of thumb" and experience is captured and shared for future use. On average 54 lessons learnt bullet points were documented per month over the eight-month period. (4) Training Presentations Periodically, as project experience is built up, training materials are developed by scouring the listservers, boards, FAQ. Monthly training presentations created by the client personnel are imparted to all vendor personnel. Presentations include all the listerv threads and their resolutions, the list of lessons learnt and pointers to any checklists or process/product document that is considered a "must read". 61

PAGE 73

CHAPTER 8 DATA ANALYSIS AND RESULTS Partial Least Squares, PLS-Graph, v3.0, Build 1126 (Chin, 1995) was used to test the research model. PLS supports the hierarchical component approach for modeling second-order factors in which the second-order factor is measured using the first-order factor scores (from the measurement model) as manifest indicators of the second order construct. In PLS, such factors can either be formative or reflective. PLS was particularly well suited for analysis of the data given its flexibility to handle second order constructs and constructs with both reflective and formative indicators. In the research model, all the first order factors (SCR, IR, ACQ, APL, CNV, TRF, PAT, VUL, PAD and CAD) are reflective. Due to limitations of the analysis tool, the SCR was measured as a single first order factor of nine items, rather than a reflective second order factor of 3 first order items representing the three dimensions of SCR pro sharing norms, generalized trust and group identification. Reflective indicators are believed to "reflect" the unobserved, underlying construct they posit to represent, with the construct giving rise to (or 'causing') the observed measures. In contrast, formative indicators have several characteristics that cogently distinguish them from reflective indicators. First, formative indicators "form" the construct as a composite (Jarvis, Mackenzie and Podsakoff, 2003). With formative indicators omitting an item is omitting a part of the construct, whereas reflective indicators are interchangeable, and the removal of an item does not change the nature of the underlying construct. Finally, reflective indicators of the same construct should show a high correlation with one another to ascertain construct validity. Formative indicators are not assumed to reflect the same underlying construct, that is, they can be independent of 62

PAGE 74

one another and measuring different factors. In fact, formative indicators of the same construct can have positive, negative, or no correlation with one another. Two models were used in the analysisa direct effects model (Figure 8.1) where SCR and IR interacted directly with KMC and a second order model (Figure 8.2), where an ISR construct was introduced to assess the combination and interaction of the two IS resourcesSCR and IR used in this study. Figure 8.1 Direct Effects Model 63

PAGE 75

Figure 8.2 Second Order Model 64

PAGE 76

The PLS modeling approach involves two steps: first, validating the measurement model and then fitting the structural model. The former is accomplished primarily by reliability and validity tests of the measurement model, followed by a test of the explanatory power of the overall model by assessing its explained variance, and the testing of the individual hypotheses (structural model). A bootstrap re-sampling procedure was conducted using 200 samples and path coefficients were re-estimated using each of these samples (Chin, 1995). In PLS, validation is done using the Composite Reliabilities (CR) and Average Variance Extracted (AVE) from the measurement model in PLS-Graph (Chin, 1998). To assess reliability and validity the block of indicator's composite reliabilities and the average variance explained (AVE) are calculated by PLS for each construct. The composite reliabilities should be greater than 0.7. The AVE measures the variance captured by the indicators relative to measurement error and it should be greater than 0.5 to justify using a construct. Moreover, the square root of each construct's AVE must be greater than the correlation of the construct to the other latent variables to ascertain discriminant reliability. The results to justify using the construct are shown in the Table 8.1 (Direct Effects model) and Table 8.2 (Second Order model) and indicate adequate composite reliabilities (CR) and AVE's. Tables 8.1 and 8.2 show the discriminant reliability of each construct. All values follow the above rule and hence discriminant reliability is exhibited by the constructs. 65

PAGE 77

O'l O'l Construct C.R. AVE Correlation of Constructs and Square root of AVE SCA lA ACQ APL CNV PAT TAF VUL SCA .883 .478 .691 lA .873 .782 .090 .889 ACQ .898 .745 .458 153 .863 APL .800 .581 518 067 .302 .762 CNV .846 .733 526 085 357 .357 .856 PAT .859 .671 .425 .124 .247 .101 .514 .819 TAF .923 .800 .679 .009 .472 .569 .487 352 .894 VUL .860 .685 .625 102 .357 .645 .395 367 .512 .828 PAD .871 .693 .590 .083 .361 .545 .374 342 .411 .465 CAD .929 .815 .037 005 .316 .056 .220 .108 .086 110 SECOND ORDER Constructs -Composite Reliabilit' AVE and Square Root of AVE Construct C.R. AVE Square Root of AVE Knowledge Management Capability (KPC) 882 .365 .604 BPO Outcome (BPOC) 837 381 .617 PAD CAD .832 .316 .903 ';} CT ii' .... () 0 .., al i c; ::I (J) () :D Ill ::I c. m 0 .., 0 ::::; (t) !l s:::: 0 c.

PAGE 78

(J) "".J Construct C.R. AVE Correlation of Constructs and Square root of AVE SCA lA ACQ APL CNV PAT TAF VUL PAD SCA .889 .484 .695 lA .875 .782 .084 .884 ACQ .898 .745 .401 .123 .863 APL .800 .581 .606 .067 .302 .762 CNV .846 .733 .420 .110 .568 .357 .856 PAT .859 .671 .410 .123 .247 .101 .513 .819 TAF .923 .800 .655 .009 .472 .569 .487 .352 .894 VUL .860 .685 .652 .102 .357 .644 .395 .367 .512 .827 PAD .871 .693 .560 .083 .362 .544 .374 .342 .411 .465 .832 CAD .929 .814 .004 .005 .187 .056 .220 .107 .086 .110 .315 SECOND ORDER Constructs -Composite Reliabilih, AVE and Square Root of AVE Construct C.R. AVE Square Root of AVE Information Systems Resources (ISR) N/A N/A N/A Knowledge Management Capability (KPC) .882 .365 .604 BPO Outcome (BPOC .837 .381 .617 ----CAD .902 -t I CT Ci' CD N (') 0 .... .... CD a 5' ::J C/l (') JJ Q) ::J a. m 0 .... (j) CD () 0 ::J a. 0 a. CD .... s:: 0 a.

PAGE 79

Estimation of Internal Consistency The survey employed multi-item scales to measure the first-order factors. The measurement properties for the reflective constructs were examined by conducting confirmatory factor analyses using PLS. To assess the internal consistency of the reflective first-order factors of SCR, KMC and BPO, coefficient alpha and composite reliability measures were used. Accordingly, as seen in Table 8.3, coefficient alpha values ranged from 0.633 to 0.883. Likewise, the composite reliabilities for all measures were high ranging from 0.800 to 0.923. Compared with coefficient alpha, which provides a lower bound estimate of internal consistency, the composite reliability is a more rigorous estimate of the reliability (Chin, 1995). The recommended levels for establishing a tolerable reliability are above the 0.70 threshold and above 0.80 for strong reliability. Consequently, evidence for internal consistency and the scales reliability are supported by these results. 68

PAGE 80

Table 8.3 Internal Consistency Measures #Items Coefficient Composite Measure Alpha Reliability IS Resources (ISR) Social Capital Resources (SCR) 9 0.856 0.883 Infrastructure Resources (IR) 2 0.849 0.875 (IR3 dropped) Knowledge Management Capability (KMC) Acquisition (ACQ) 3 0.831 0.898 Protection (PAT) 3 0.764 0.859 Application (APL) 3 0.650 0.800 Transfer (TRF) 3 0.875 0.923 Conversion (CNV) 2 0.633 0.846 (CNV3 dropped) BPO Outcomes (BPO) Vulnerability Costs (VUL) 3 0.763 0.860 Production Costs (PAD) 3 0.779 0.871 Coordination Costs (CAD) 3 0.883 0.929 69

PAGE 81

Dimensionality, Convergent, and Discriminant Validity In the direct effects model (Figure 8.1 ), the SCR and lA constructs are conceptualized to have direct relationships with KMC by testing the direct effects of each of the first-order factors on KMC, whereas the second order model suggests an explicit multidimensional structure (ISR) whereby SCR, and IR combine to form an emergent force (ISR) that in turn can affect KMC. As indicated by the results in Table 8.4, all the loadings were statistically significant based on t-statistics generated from running a bootstrap on the data, except for one (BPO CRD). To model the complex character of a firm's ISR, which was operationalized effectively in a formative way by a composite across different, unique sources of the constructSCR and IR. As a result, a second order model was needed to understand and define how firms use ISR as a combination of SCR and lA. 70

PAGE 82

Table 8.4 Structural Model Results Path Direct Effects Second Order Model Model Structural Model Hypothesized Relationships SCR 7 KMC (H1 ) 0.809 {13.479) N/A IR 7 KMC (H2) 0.037 (0.285) N/A ISR 7 KMC (H3) N/A 0.732 {7.952) KMC 7 BPO (H4 ) 0.640 (5.045) 0.654 (3.670) ISR x KMC 7 BPO (H5 ) N/A 0.0232 (4.579) Measurement Model SCR N/A 0.989 (9.814) ISR IR N/A 0.180 (1.974)*"* KMC 7ACQ 0.725 (7.049) 0.724 (6.795) KMC 7 CNV 0.776 (10.243) 0.775 (10.502) KMC 7 APL 0.663 (6.670) 0.664 (6.599) KMC 7 PAT 0.564 (1.812) **"* 0.566 (1.899) **"* KMC 7 TRF 0.850 (13.045) 0.852 (16.975) BPO 7 VUL 0.762 (4.605) 0.752 (4.244) BPO 7 PAD 0.833 (12.427) 0.851 (11.181) BPO 7 CAD 0.451 (1.531) 0.571 (1.595) Dependent Variables KMC 0.6606 0.5360 BPO 0.4272 0.4275 BPO w/ Moderator Term N/A 0.4507 p<0.001; ** p<0.01; *"* p<0.05; **"*p<0.1 0 Notes: Parameter estimates are standardized with t-values 71

PAGE 83

Second-Order Model As shown in Figure 8.2, the second order model specification hypothesizes ISR as a second-order formative construct formed by two first-order factors made up of SCR and IR. Formative constructs are conceived to be 'caused' by the underlying measurement items where each lower-order item represents a distinct contribution to the higher-order latent construct (Jarvis, Mackenzie and Podsakoff, 2003). Although it was possible that the factors might be highly interrelated enough to be treated as reflective factors, this was not supported by the results. In a formative model, SCR and IR may be, but need not, be highly correlated, and these results suggest this they are not highly correlated (Table 8.5). Table 8.5 Correlation Matrix Factors in ISR Second Order Model I I SCR IR SCR Pearson Correlation 1 .000 Sig. (2-tailed) 1.000 N 119 119 IR Pearson Correlation .000 1 Sig. (2-tailed) 1.000 N 119 119 A reflective model would have very high correlations among the first-order factors (greater than 0.70), which was not the case. Thus a formative model seems more likely for ISR. Logically, a change in SCR, for example, does not necessarily imply an equal change in IR, hence a reflective model is less likely. Moreover, the effective use of the resources is likely to change over time and be affected in a different way by other factors. 72

PAGE 84

As such, the sub-constructs of ISR form a higher-order formative model that most accurately and parsimoniously captures the multi-dimensional nature of ISR. Before testing to determine if the second-order formative model of ISR (Figure 8.1) is a better fit to the data than the first-order factor model (Figure 8.2), model's statistics for reliability and convergent and discriminant validity was used to verify the measurement models. The results for the two models are shown in Tables 8.1 and 8.2. These metrics were similar and provided support for reliability and convergent and discriminant validity for all constructs. Hypothesis Testing The adequacy of the psychometric properties in the measurement model allowed further testing the direct effect hypotheses (H1 and H 2 ) using the direct effects model depicted in Figure 8.1 As shown in Table 8.4 all first-order factors of KMC were significantly related with knowledge management capabilities, KMC as well as all first order factors of BPO were significantly linked with the higher order BPO construct, except CAD. While no minimum threshold value for weights has been established, the statistical significance of the weights can be used to determine the relative importance of the indicators in forming the construct. As the interpretation of the weights is similar to the beta coefficients in a standard regression model, it is usual to have lower absolute weights as compared to loadings. For hypothesis, H1 the results show that the effect of social capital resources, SCR = 0.809; t = 13.479; p < 0.001) was a lot more important than the effect of Infrastructure resources, IR = 0.037; t = 0.285; p < 0.05) on the knowledge management capabilities construct, KMC. Hypothesis H1 is therefore supported by the data and SCR has a significant influence on KMC. However, hypothesis H2 is not 73

PAGE 85

supported by the data; that is Infrastructure resources (IR) on its own does not have a significant influence on KMC (13 = 0.037; t = 0.285; p > 0.05). The direct effect model also was used to test the hypothesis H4 It is seen that KMC has a significant positive influence on BPO Outcomes (13 = 0.640; t = 5.045; p < 0.001). The second-order factor model with the higher order construct, ISR is used to test the remaining two hypothesis, H3 and H5 To estimate the hypothesized second order formative model of ISR from the two first order factors of Social capital resources (SCR) and Infrastructure resources (IR), the coefficients for each of the two first order factors were modeled using a principal components factor analysis, following the procedures in Karimi, Somers and Bhattacharjee (2007b) and Diamantopoulos and Winklhofer (2001). The results of the factor analysis are shown in Tables 8.6 and 8.7. 74

PAGE 86

Table 8.6 Principal Components Analysis for Second Order Model for ISR C()IT1P[nen_t 1 2 IR1 IR2 PSN ID GT -.058 .078 .871 .810 .828 Eigenvalue 2.115 % Variance Explained 42.304 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Table 8.7 Component Score Coefficient Matrix for ISR IR1 IR2 PSN ID GT Com -----------------1 -.038 .026 .413 .384 .392 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. 75 anent ----------2 .933 .932 .007 -.018 .037 1.736 34.718 -----.537 .535 -.006 -.019 .012

PAGE 87

The correlations among the two first order factors is very low, demonstrating that the content captured by the first order factors are distinct from one another. The coefficients of the first order factors are statistically significant, providing justification for the existence of the hypothesized formative second order model of ISR. It is seen that the second order model is a more parsimonious model with fewer parameters to be estimated. The structural link from ISR to KMC (see Table 8.4) is positive and significant (13 = 0. 732; t = 7.952; p < 0.001 ). The results provide empirical support for H3 and validates the expectation that ISR as a set of resources has a significant association with building KMC. The predictive power of the models for the BPO outcomes construct is shown by the R2 values in Table 8.4. The final hypothesis concerns the presence of an interaction between ISR and KMC with BPO (H5), which states that ISR will moderate the relationships between KMC and BPO. The interaction term (ISR x KMC) was formed by cross-multiplying all standardized items of each constructs, following the procedure in Chin, Malcolm and Newsted (2003) The results suggest an assembly of social capital resources and infrastructure resources can slightly strengthen the influence building knowledge capabilities has on business process outcomes (13 = 0.654; t=3.670; p < 0.001) and supports H5 Comparing the Moderator Model with Second Order Model As shown in Figure 8.2, to assess the interaction of ISR and KMC on BPO the two estimated models direct effect and second order models were compared in two steps to judge the incremental variance explained by adding the moderating effect (Chin, Malcolm and Newsted, 2003). The second order model was chosen to add the moderating effect to, since the second order model produced a higher A-square for the BPO outcomes measure. 76

PAGE 88

The R2 for BPO outcomes was 0.4275 for the second order model versus 0.4272 for the direct effects model (see Table 8.4). The test for the moderated relationship was conducted by using f1R2 to draw conclusions about the moderator effect size since "the use of the path coefficient of an interaction term will lead to spurious conclusions" (Carte and Russell, 2003). This approach prevents an incorrect interpretation of the significance of the interaction term when it is correlated with its constituent parts, that is, its main effect (Karimi, Somers and Bhattacharjee, 2007a). The change in R2 was examined by comparing the results of two models derived from the second order model (which was chosen as it has the higher R2 compared to the direct effect model See Table 8.4). The R2 for the BPO construct from these models are used to calculate the moderating effect of ISR relationship between KMC and BPO. The relationship between BPO and KMC can be specified as follows: Base model: BPO = f (KMC, E ... ), ( 1) and compared to the following model that includes the relationship of KMC and the interaction of KMC and ISR: BPO = f (KMC, KMCISR, E ... ), (2) or more specifically by an equation1 of the form: (3) 1 The sensitivity of results to the inclusion of an interaction term is often taken as a sign of multicollinearity. If there is high multicollinearity, it can lead to large standard errors on the model parameters. However, and more importantly, we are not directly interested in the significance or insignificance of the model parameters per se anyway. Instead, we are interested in the marginal effect of X on Y. In the case of Eq. (3), this is: (')Y/(')KMC = 131 + I32KMCISR (Karimi, Somers, Bhattacharjee, 2007a) 77

PAGE 89

Consequently, denotes the significance of the interaction term when added to the base model. The standardized path estimate from the product construct (ISR x KMC) to BPO indicates how a change in ISR would change the influence of KMC on the dependent construct BPO. The A-square for the second order model with the interaction term (R2 = 0.4507) was compared with the A-square for the second order model without the interaction term (R2 = 0.4275) to assess the strength of the moderating effect. The true effect of the interaction term was calculated through the effect size2 (f 2 ) where 0.02, 0.15, and 0.35 has been suggested as small, moderate, and large effects, respectively (Cohen, 1988). The f statistics, which is based on the differences in R2 between the two models, was determined and used to compute the pseudo-F statistic3 (Carte and Russell, 2003; Chin, Malcolm and Newsted, 2003). The interaction effect from ISR x KMC on BPO produces a small effect size (f 2= 0.023) thereby supporting the posited moderating effects of ISR on the relationship between KMC and BPO. An F statistic that is significantly greater than 1.00 leads to the rejection of H0 : = 0. The Pseudo F-statistic was calculated as F=4.519, thereby providing statistical significance to hypothesis, The analysis suggests that the additional variance explained by introducing the moderator, ISR X KMC significantly adds to the variance explained in BPO (R2 increasing from .4275 to .4507 in the second order model when the moderator is included). 2 Effect size f2 = [R2 (included) -R2 (excluded)] I [(1R2 (included)] 3 The pseudo-F statistic is computed using the formula f (n-k-1), with 1, (n-k) degrees of freedom where n is the sample size and k is the number of constructs in the model (Chin, Malcolm and Newsted, 2003). 78

PAGE 90

Consequently, it is evident that ISR and KMC demonstrate a mutually reinforcing complementarity relationship (Zhu, 2004). ISR and KMC jointly improve BPO. Therefore, the impact of KMC on BPO will be contingent on the level of IS resources; the impact will be stronger for firms with the higher levels of IS resources in comparison to the firms with the lower levels. 79

PAGE 91

CHAPTER 9 DISCUSSION AND CONCLUSIONS Three research questions were examined in this study: (1) Can IS resources particularly social capital resources from social capital theory translate to improved knowledge management capabilities between the client and vendor in the offshore BPO arrangement? (2) Do knowledge management capabilities impact the BPO outcomes? (3) Do IS resources and knowledge management capabilities have a complementary relationship with BPO outcomes? To answer the questions, five research hypotheses were tested. Using survey data from a field study of a technical support organization that deployed BPO, significant evidence was found for the direct association of social capital resources (H1 ) and building knowledge management capabilities, but not for the direct association of infrastructure resources (H2 ) on knowledge management capabilities. From a practical perspective, however, infrastructure resources are also important for both building knowledge management capabilities and this is supported in the second order model where IS resources (ISR) as a combination of social capital resources (SCR) and infrastructure resources (IR) was significantly related to knowledge management capabilities (H3). The model also supported the relationship between knowledge management capability on BPO outcomes (H4). Finally the complementary effect of IS resources on BPO outcome through KM capability building was also supported (H5). These findings demonstrate that the Direct Effects model was not sufficient for explaining how firms build knowledge management capabilities. There is an additive effect of the set of IS resources (ISR) on building knowledge management capabilities where the effect of each resource 80

PAGE 92

seemingly depends upon commingling with others. The study also found that the co presence of IS resources tends to supersede the direct effects of each IS resource alone. Integrated together, they result in a performance-enhancing resource bundle for building knowledge management capabilities. Therefore, the individual dimensions of IS resources SCR and IR should not be considered in isolation from each other, but should be treated in a collective and mutually reinforcing manner. This finding is similar to those reported in recent studies (Karimi, Somers and Bhattacharjee, 2007a; Tanriverdi, 2005; Tanriverdi and Venkataraman, 2005). The co-presence of resources provides unique value to the firm. Moreover, the synergies arising from the co-presence of resources are much more difficult to observe and imitate by other firms (Tanriverdi, 2005). Regarding the second question, the results did provide compelling evidence that building knowledge management capabilities has positive association with BPO outcomes. Also, the association of building knowledge management capabilities with BPO outcomes is contingent on the co-presence of IS resources, which supports the third research question as well. The importance of the co-presence of IS resources was demonstrated by their addition to the model. When included, it increased the explained variance in business process outcomes to 45 percent (R2 = .4507), and had a small effect size (f 2 of 0.0232). This confirmed the hypothesis that IS resources can intervene to strengthen the relationship between building knowledge management capabilities and business process outsourcing outcomes. The notion of complementarities suggests that firms that possess a wide range of resources are better able to exploit the synergistic benefits of these resources than those that possess fewer resources or a lesser level of each resource (Tanriverdi and Venkatraman, 2005). This is important since competitors 81

PAGE 93

usually lack the strategic foresight to recognize complementarities. Even, if they recognize the complementarities, to imitate them successfully, they have to make simultaneous changes in IT infrastructure (Karimi, Somers and Bhattacharjee, 2007a). The client vendor relationship in a BPO can be quite arduous, with jobs at stake and differences in capabilities of the two organizations. This study shows that if social capital resource is increased in the BPO relationship, then the overall outcome of the BPO can be improved. With the development of shared work practices, social capital resource is created resulting in improved knowledge management processes. As a result the BPO outcome is improved. Practitioners may note that the investment into building successful working level relationships in a BPO can indeed be worthwhile. Drawing on the resource based view of the firm, this research demonstrates that IS resources particularly infrastructure and relationship resources can influence knowledge management capabilities and improve BPO outcomes. The complementarity effect of IS resources and knowledge management capabilities on BPO outcome implies that investments into building knowledge management capabilities together with building the IS resources in conjunction leads to higher BPO outcome gains. Implications for Research A research model based on RBV to study the contribution of social capital resource as an IS resource on the knowledge management capabilities in a BPO is developed and validated. By defining the social capital resource and measuring it in the BPO setting this study makes a valuable contribution to the body of IS research. The results of this study support the research background that benefits of social capital resource can be seen in knowledge management capabilities. The results indicate that the BPO outcome is improved. Researchers have been looking for the impact of social 82

PAGE 94

capital resources in downstream work processes and this study shows that social capital resources can improve outcomes by improving knowledge management capabilities. A measurement of BPO outcome based on Coordination theory is also developed and validated. This BPO measurement model may be used in future research to study other BPO scenarios as well and measure the performance outcomes of a BPO. Additionally, the study can be extended to other industries with knowledge intensive processes such as the healthcare industry as more processes in that industry are outsourced (Scott and Ghosh, 2005b). Implications for Practice The results support the need for building social capital resources in offshore BPO and deploying IT infrastructure resources together with social capital resources to exploit the complementarity effect. Several papers have studied the enablers for social capital and management practices that can help foster outsourcing success as well as generate social capital (Rottman and Lacity, 2004). The organizational enabling conditions stated for supporting social capital creation fall into three categories: (1) Bridging, where individuals are brought together purposely for collective work, (2) Bonding, where cognitive norms and implicit understanding is developed by personnel on both sides, and (3) Linking, where structural connections are established for jointly owning ongoing activities (Kowch, 2005). Organizational processes that involve resources on both sides of the relationship can be social capital enablers, if such processes posses the above characteristics. Examples of these organizational practices may be: (1) creating opportunities for exchange, (2) creating an expectation that such combinations and exchanges will have value, (3) creating motivation for both sides to participate and (4) creating structural 83

PAGE 95

norms and symmetries to support combination capability (Nahapiet and Ghoshal, 1998). Wenger (1998) found factors and their inter-relatedness to significantly impact social capital formation and knowledge transfer in communities of practice. Several authors have argued that social capital creation is possible through pursuit of organizational enabling activities (O'Brien, Phillips and Patsiorkovsky, 2005; Schmid, 2000). Delone, et. al. (2005) describes the differences between bridging, bonding and linking activities in social capital creation. They argue that any mechanism that establishes repetitive and routine aspects of tasks in the inter-organizational relationship can be classified as "bridging". Organizational activities that help organizations and teams interact and coordinate implicitly by building inner expectations, through cognition mechanisms of norms, beliefs and trust can be referred to as "bonding" activities. Finally, Baum and Ziersch (2006) describe "linking" activities as organizational practices aimed at building vertical connections, spanning difference of power structure to help reduce inequalities and build common responsibility for tasks in the inter-organizational relationship. Generalizability of the Results The data collection from a single organization can raise questions about the generalizability of any theory. However it is usually recommended to replicate any study in the new setting to justify porting any results or theory from the research setting to a new setting (Lee and Baskerville, 2003). While the above approach to generalizing results may be too restrictive, the authors go on to say that even if a 100 or 1000 organizations were included in the study, it would only extend generalizability to those 1 00 or 1 000 organizations that fell in the scope of the study setting and not guarantee 84

PAGE 96

universal generalizability. This undermines the need to include more than one organization in a confirmatory research study to support the generalizability of the results from the study. The setting of this study was a large multinational customer services organization that supports multi vendor networking equipment. The organization had involved an offshore vendor organization to outsource some of the transactions in the service processes. The processes by their description are fairly representative of the industry and are more or less universally followed in delivering product support and network design work through out the telecommunications and high technology industry by many other organizations. Thus there are no peculiarities with these processes, or the transactions that these processes support or the personnel involved in the processes. Hence, there is no reason to believe that these results and the confirmed theory cannot be applied to other customer service outsourcing settings or possibly other BPO settings as well. Moreover, customer service and technical product support accounts for a major percentage of offshore BPO at greater than 30% of all outsourced processes (Segal, 2003). Hence the results and the confirmed theory are of particular interest to the IS community and do have general appeal. Limitations The limitations of this research are the fact that only one organization was surveyed. Clearly these two processes for the selected organization may not be representative of the general customer service processes that are being outsourced. However, the descriptions of the processes do not indicate any particular peculiarities that may render them so unique that other customer service processes in other organizations would be grossly different. 85

PAGE 97

The IS resources construct in the research model included relationship resources in the form of social capital resources and infrastructure resources. A third form of IS resource often used in RBV studies are knowledge resources (Karimi, Somers and Bhattacharjee, 2007a). A firm's knowledge resources are the unique skills, expertise, insights, experience, and intellectual capital that it uses for building a capability. Knowledge exploitation requires sharing relevant knowledge among members of a firm to promote mutual understanding and comprehension. While the types and forms of knowledge resources for the organization were catalogued in the data collections section, this study did not however, incorporate a measurement for knowledge resources in the research model. This was a decision based on confusion regarding where such a measurement may have fit in the model -(1) in the outcomes measure as knowledge resources could be considered units of coordination information or (2) in the resources measure if knowledge resources were considered to be a resource for this study. It is a fact that adequate and high quality knowledge resources are needed to facilitate the effectiveness of knowledge management capabilities. If workers could not locate or exploit these knowledge resources from the KMS, why would they increase their use of the knowledge management capability? However, it can be safely assumed that there were indeed adequate knowledge resources in the KMS. This was a precondition of the research, since these service processes have been operational in the client organization for many years using the same knowledge management systems and tools. The processes were outsourced partially to an offshore vendor to increase staffing at a lower labor rate. The knowledge resources that were existing in the KMS were still relevant after the BPO implementation and hence form a resource base that is accessible by both client and vendor staff. Given that the study has collected the detailed records of how 86

PAGE 98

many knowledge entries were added over a 8 month duration in the various knowledge management tools, it is also quite feasible to develop a measure of knowledge resources and incorporate into the model and retest the hypotheses for any changes in the results as a future exercise. 87

PAGE 99

APPENDIX AFINAL SURVEY QUESTIONS Strongly Disagree 1 Disagree -2 Neutral3 Agree-4 Strongly Agree -5 1 What Outsourced process are you answering the questions a. Customer Support with respect to? b. Network Consulting 2. Years of on the job experience? 3. Years of schooling (K-12 counts as 13years) 4. Gender? a. Male b. Female There is a norm of teamwork among staff in the BPO 1 2 3 4 5 relationship. There is a norm of openness of diverse/conflicting views among 1 2 3 4 5 staff in the BPO relationship There is a norm of tolerance of mistakes among staff in the 1 2 3 4 5 BPO relationship Client and vendor staff in the BPO relationship share common 1 2 3 4 5 values. Client and vendor staff in the BPO relationship share common 1 2 3 4 5 goals Client and vendor staff in the BPO relationship have strong 1 2 3 4 5 group identification Client and Vendor staff in the BPO relationship give credit to 1 2 3 4 5 people where it is due Obligations between organizations are always sustained by 1 2 3 4 5 staff in the BPO relationship Client and Vendor staff in the BPO relationship use other's 1 2 3 4 5 knowledge appropriately Our organizations have processes for acquiring knowledge 1 2 3 4 5 from our BPO partners Our organizations have processes for generating new 1 2 3 4 5 knowledge from existing knowledge in the BPO relationship Our organizations have processes for exchanging knowledge 1 2 3 4 5 between individuals in the BPO relationship. Our organizations have processes for converting knowledge 1 2 3 4 5 into the design of future BPO support services Our organizations have processes for distributing knowledge 1 2 3 4 5 throughout the two organizations in the BPO relationship Our organizations have processes for integrating different 1 2 3 4 5 sources and types of knowledge 88

PAGE 100

Our organizations have processes for applying knowledge 1 2 3 4 5 learned from experience Our organizations have processes to make knowledge 1 2 3 4 5 accessible to those who need it Our organizations have processes to quickly link sources of 1 2 3 4 5 knowledge in solving problems I am encouraged to discuss my work with employees in other 1 2 3 4 5 workgroups Benefits of sharing knowledge in my organization outweigh the 1 2 3 4 5 costs Knowledge is freely exchanged freely through email, 1 2 3 4 5 presentations, newsletters, etc. Our organizations have processes to protect knowledge from 1 2 3 4 5 inappropriate access in the BPO relationship Our organizations have processes to protect knowledge from 1 2 3 4 5 inappropriate use in the BPO relationship Our organizations clearly communicate the importance of 1 2 3 4 5 protecting knowledge After the BPO, the my organization adapts more quickly to 1 2 3 4 5 unanticipated changes After the BPO, the my organization can react more quickly to 1 2 3 4 5 new market info After the BPO, my organization's response to market changes 1 2 3 4 5 involves a less lengthy process After the BPO, the transaction costs related to these 1 2 3 4 5 outsourced processes are lower After the BPO, we can serve our customers more effectively 1 2 3 4 5 now After the BPO, less resources are required to operate the 1 2 3 4 5 outsourced processes After the BPO, a decreased number of resources are needed 1 2 3 4 5 to support the outsourced process After the BPO, a decreased number of managerial contacts are 1 2 3 4 5 needed to support the outsourced process After the BPO, less time is required to get status and updates 1 2 3 4 5 about the outsourced process Appropriate hardware, software, and network infrastructures 1 2 3 4 5 are in place to support the BPO Appropriate knowledge systems and tools are in place in the 1 2 3 4 5 infrastructure to support the BPO Access technologies are administered in the infrastructure to 1 2 3 4 5 allow appropriate access to knowledge 89

PAGE 101

APPENDIX BOUTER MODEL LOADINGS (DIRECT EFFECTS MODEL) Original Mean of Standard T -Statistic sample subsamples error estimate PAD (Composite Reliability = 0.871, AVE= 0.693) PRD1 0.8024 0.7640 0.1806 4.4442 PRD2 0.7965 0.7748 0.0995 8.0081 PRD3 0.8955 0.8940 0.0495 18.0820 BPO (Composite Reliability = 0.837, AVE= 0.381 ) VUL1 0.7612 0.7522 0.1178 6.4632 VUL2 0.7034 0.6950 0.1496 4.7023 VUL3 0.2575 0.2452 0.2886 0.8923 PRD1 0.7027 0.6226 0.2374 2.9595 PRD2 0.5767 0.5489 0.1527 3.7766 PRD3 0.8169 0.7862 0.0924 8.8370 CRD1 0.5874 0.4700 0.3836 1.5311 CRD2 0.4319 0.3375 0.3956 1.0917 CRD3 0.5102 0.4070 0.3956 1.2898 VUL (Composite Reliability = 0.860, AVE= 0.685) VUL1 0.9417 0.9242 0.1240 7.5914 VUL2 0.9472 0.9220 0.1072 8.8329 VUL3 0.5203 0.4549 0.3004 1.7322 CAD (Composite Reliability = 0.929, AVE= 0.815) CRD1 0.9337 0.9201 0.0611 15.2745 CRD2 0.8529 0.8602 0.0768 11.1064 CRD3 0.9189 0.9142 0.0734 12.5191 SCR (Composite Reliability = 0.883, AVE= 0.478) PSN1 0.9048 0.9014 0.0410 22.0734 PSN2 0.7252 0.7005 0.1223 5.9303 PSN3 0.9000 0.8992 0.0416 21.6358 ID1 0.6459 0.6515 0.1439 4.4886 ID2 0.2014 0.1757 0.3001 0.6711 ID3 0.6882 0.6968 0.1125 6.1178 GT1 0.5527 0.5441 0.1418 3.8980 GT2 0.6958 0.6823 0.1343 5.1803 GT3 0.6518 0.6057 0.1434 4.5468 90

PAGE 102

IR (Composite Reliability = 0.875, AVE= 0.782) IR1 0.7509 0.8204 0.2407 3.1202 IR2 0.9999 0.8303 0.2635 3.7951 CNV (Composite Reliability = 0.846, AVE= 0.733) CNV1 0.8332 0.7903 0.2135 3.9020 CNV2 0.8788 0.8903 0.0964 9.1142 KMC (Composite Reliability = 0.882, AVE= 0.365) ACQ1 0.5561 0.5152 0.1683 3.3038 ACQ2 0.7731 0.7527 0.0964 8.0238 ACQ3 0.4839 0.4484 0.1548 3.1252 CNV1 0.6134 0.5767 0.2150 2.8535 CNV2 0.7109 0.7197 0.1093 6.5067 APL1 0.5452 0.5424 0.1626 3.3526 APL2 0.2425 0.2644 0.2303 1.0530 APL3 0.6265 0.6312 0.1150 5.4466 PRT1 0.5493 0.5208 0.2569 2.1381 PRT2 0.2788 0.2665 0.2985 0.9341 PRT3 0.4824 0.4770 0.2581 1.8692 TRF1 0.7545 0.7270 0.1117 6.7527 TRF2 0.6901 0.6858 0.1239 5.5702 TRF3 0.8269 0.8094 0.0757 10.9251 APL (Composite Reliability = 0.800, AVE= 0.581 ) APL1 0.8358 0.8069 0.1053 7.9361 APL2 0.5481 0.5182 0.2736 2.0033 APL3 0.8617 0.8529 0.0597 14.4405 ACQ (Composite Reliability = 0.898, AVE= 0.745) ACQ1 0.8557 0.8330 0.0824 10.3892 ACQ2 0.8958 0.9023 0.0335 26.7133 ACQ3 0.8372 0.8183 0.0896 9.3450 PAT (Composite Reliability = 0.859, AVE= 0.671) PRT1 0.8647 0.8184 0.1562 5.5350 PRT2 0.7758 0.7131 0.2472 3.1382 PRT3 0.8151 0.7745 0.1671 4.8772 TRF (Composite Reliability = 0.923, AVE= 0.800) TRF1 0.9037 0.8895 0.0578 15.6432 91

PAGE 103

TRF2 TRF3 0.8578 0.9203 0.8565 0.9199 92 0.0603 0.0263 14.2147 34.9735

PAGE 104

APPENDIX C-OUTER MODEL LOADINGS (SECOND ORDER MODEL) Original Mean of Standard T -Statistic sample subsamples error estimate PAD (Composite Reliability = 0.871 AVE= 0.693) PRD1 0.8024 0.7870 0.1376 5.8332 PRD2 0.7966 0.7644 0.1162 6.8528 PRD3 0.8955 0.8750 0.1141 7.8501 BPO (Composite Reliability = 0.837, AVE= 0.381 ) VUL1 0.7614 0.7375 0.1817 4.1896 VUL2 0.7036 0.6759 0.1914 3.6756 VUL3 0.2577 0.2609 0.3048 0.8456 PRD1 0.7026 0.6449 0.1925 3.6490 PRD2 0.5769 0.5369 0.1528 3.7748 PRD3 0.8170 0.7642 0.1474 5.5429 CRD1 0.5871 0.5093 0.3360 1.7474 CRD2 0.4316 0.3638 0.3142 1.3737 CRD3 0.5098 0.4683 0.3307 1.5416 VUL (Composite Reliability = 0.860, AVE= 0.685) VUL1 0.9417 0.9320 0.0375 25.0945 VUL2 0.9472 0.9308 0.0412 22.9706 VUL3 0.5203 0.4676 0.3168 1.6424 CAD (Composite Reliability = 0.929, AVE= 0.814) CRD1 0.9337 0.9297 0.0411 22.7184 CRD2 0.8528 0.8365 0.0825 10.3371 CRD3 0.9189 0.9234 0.0447 20.5417 SCR (Composite Reliability = 0.889, AVE= 0.484) PSN1 0.8584 0.8613 0.0558 15.3908 PSN2 0.7402 0.7358 0.0815 9.0833 PSN3 0.8577 0.8634 0.0373 22.9875 ID1 0.6345 0.6255 0.1286 4.9343 ID2 0.3459 0.3443 0.2701 1.2806 ID3 0.6595 0.6647 0.1282 5.1431 GT1 0.5456 0.5248 0.1614 3.3813 GT2 0.7236 0.7201 0.0966 7.4898 GT3 0.7462 0.7321 0.1035 7.2113 93

PAGE 105

IR (Composite Reliability = 0.875, AVE= 0.782) IR1 0.7505 0.8792 0.2219 3.3827 IR2 0.9999 0.8852 0.1208 8.2806 CNV (Composite Reliability = 0.846, AVE= 0.733) CNV1 0.8332 0.7733 0.1972 4.2242 CNV2 0.8788 0.9025 .0571 15.3948 KMC (Composite Reliability = 0.882, AVE= 0.365) ACQ1 0.5542 0.5107 0.1558 3.5572 ACQ2 0.7717 0.7572 0.1062 7.2670 ACQ3 0.4810 0.4317 0.1624 2.9620 CNV1 0.6109 0.5423 0.2221 2.7506 CNV2 0.7079 0.7277 0.0880 8.0461 APL1 0.5465 0.5501 0.1723 3.1724 APL2 0.2433 0.2797 0.2371 1.0261 APL3 0.6258 0.6206 0.1389 4.5041 PRT1 0.5487 0.4938 0.2566 2.1382 PRT2 0.2820 0.2531 0.3070 0.9185 PRT3 0.4862 0.4817 0.2178 2.2323 TRF1 0.7559 0.7540 0.0825 9.1645 TRF2 0.6923 0.7132 0.0974 7.1083 TRF3 0.8283 0.8195 0.0687 12.0641 APL (Composite Reliability = 0.800, AVE= 0.581 ) APL1 0.8362 0.8099 0.1146 7.2956 APL2 0.5485 0.5325 0.2711 2.0235 APL3 0.8612 0.8426 0.0828 10.3954 ACQ (Composite Reliability = 0.898, AVE= 0.745) ACQ1 0.8556 0.8254 0.0769 11.1326 ACQ2 0.8961 0.9068 0.0289 30.9539 ACQ3 0.8368 0.8048 0.1088 7.6883 PAT (Composite Reliability = 0.860, AVE= 0.671 ) PRT1 0.8635 0.8135 0.1507 5.7315 PRT2 0.7765 0.7063 0.2568 3.0236 PRT3 0.8160 0.8021 0.1615 5.0539 TRF (Composite Reliability = 0.923, AVE= 0.800) TRF1 0.9036 0.9014 0.0401 22.5501 94

PAGE 106

TRF2 TRF3 0.8579 0.9202 0.8614 0.9210 95 0.0552 0.0266 15.5376 34.6558

PAGE 107

BIBLIOGRAPHY Adler, P.S. and Kwon, S.W. (2002). Social Capital: Prospects for a New Concept. Academy of Management Review, 27(1), 17-40. Alavi, M. and Leidner, D.E. (2001). Review: Knowledge management and knowledge management systems: conceptual foundations and research issues. MIS Quarterly, 25(1), 107-136. Andreu, R., and C. Ciborra. (1999). Organizational Learning and Core Capabilities Development: The Role of IT. Journal of Strategic Information Systems, 5, 111-127. Ang, S. and Slaughter, S. A. (1998). Organizational Psychology and Performance in IS Employment and Systems Outsourcing. Proceedings of the 31nd Annual Hawaii International Conference on System Sciences (635-643). Honolulu, HI. Ang, S. and Straub, D. (1998). Production and Transaction Economies and IS Outsourcing. MIS Quarterly, 22(4), 535-552. Argote, L. (1999). Organizational Learning: Creating, Retaining and Transferring Knowledge. Boston: Kluwer Academic. Aron, R., Clemons, E.K. and Reddi, S. (2005). Just Right Outsourcing: Understanding and Managing Risk. Journal of Management Information Systems, 22(2), 37-55. Aubert, B. A., Dussault, S. Patry, M. and Rivard, S. (1999). Managing the Risk of IT Outsourcing. Proceedings of the Annual Hawaii International Conference on System Sciences. Honolulu, HI. Baum, F.E.., Ziersch, A.M., E.K (2003). Social Capital. Journal of Epidermiology and Community Health, 57, 320-323. Barney, J. B. (1991 ). Firm Resources and Sustained Competitive Advantage. Journal of Management, 17(1), 99-120. Barney, J. B., Wright, M. and Ketchen, D.J. (2001 ). The resource based view of the firm: Ten years after 1991. Journal of Management, 27(3), 625-641. Becerra-Fernandez, I. and Sabherwal, R. (2001). Organizational knowledge management: A contingency perspective. Journal of Management Information Systems, 18(1), 23-56. Bharadwaj, A. (2001). A Resource-based Perspective on Information Technology Capability and Firm Performance: An Empirical Investigation. MIS Quarterly, 24(1), 169-196. Brewer, G. A. (2003). Building Social Capital: Civic Attitudes and Behavior of Public Servants. Journal of Public Administration Research and Theory, 13, 5-26. 96

PAGE 108

Broadbent, M., Weill, P.,and St. Clair, D (1999). The Implications of Information Technology Infrastructure for Business Process Redesign. MIS Quarterly, 23{2), 159-182. Brown, J.S. and Duguid, P. (1991). Organization Learning and Communities of Practice: Toward a unified view of working, learning and innovation. Organization Science, 2(2), 40-57. Brown, J.S. and Duguid, P. (2001). Knowledge and Organizations: A Social-Practice Perspective. Organization Science, 12(2), 50-66. Burgess, D (2005) What Motivates Employees to Transfer Knowledge outside their work unit? Journal of Business Communications, 42(4), 324-348 Carte, T.A, and Russell, C.J. (2003). In Pursuit of Moderation: Nine Common Errors and Their Solutions. MIS Quarterly, 27(3), 479-501. Carmel, E. and Agarwal, R. (2002). The Maturation of Offshore Sourcing of Information Technology Work. MIS Quarterly Executive, 1(2), 65-76. Carroll, M. C. and Stanfield, J.R. (2003). Social capital, Karl Polanyi, and American social and institutional economics. Journal of Economic Issues. 37, 397-404. Chan, Y. (2002) Why Haven't We Mastered Alignment? The Importance of the Informal Organizational Structure. MIS Quarterly Executive, 1(2), 97-112. Chen, S. and Choi, C.J. (2005). A Social Exchange Perspective on Business Ethics: An Application to Knowledge Exchange. Journal of Business Ethics, 62, 1-11. Cheung, P.K. and Kwan, M.M. (2004). Towards a Unified Framework on Knowledge Management: An Organizational Knowledge Management Perspective. Proceedings of the Tenth Americas Conference on Information Systems (2295-2304). New York, NY. Chin, W.W (1995) Partial Least Squares is to Lisrel as Principal Components Analysis is to Common Factor Analysis. Technology Studies, 2, 315-319. Chin, W.W. (1998). Issues and Opinions on Structural Equation Modeling. MIS Quarterly, 22{1), 7-16. Chin, W.W., Malcolm, B.L and Newsted, P.R. (2003). A Partial Least Squares Latent Variable Modeling Approach for Measuring Interaction Effects: Results from a Monte Carlo Simulation Study and an Electronic-Mail Emotion/Adoption Study Information Systems Research, 14(2) 189-217 Clark, T. D Zmud, R. W. and McGray, G. E (1995). The Outsourcing of Information Services: Transforming the Nature of Business in the Information Industry. Journal of Information Technology, 10, 221-237. 97

PAGE 109

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates. Cohen, W.M., and Levinthal, D.A. (1990). Absorptive capacity: a new perspective on learning and innovation. Administrative Science Quarterly, 35, 128-152. Coleman, J. S. (1988). Social Capital in the Creation of Human Capital. The American Journal of Sociology, 94, 95. Cropanzano, R. and Mitchell, M.S. (2005). Social Exchange Theory: An Interdisciplinary Review. Journal of Management, 31 (6), 874-900. Cross, J. (1995). IT Outsourcing: British Petroleum. Harvard Business Review, 5, 94-102. Davenport, T.E., Prusak, L. (1998). Working Knowledge: How organizations manage what they know. Boston: Harvard Business School Press. Delane, W., Espinosa, J.A., Lee, G. and Carmel, E. (2005). Bridging Global Boundaries for IS Project Success. Proceedings of the 3!' Hawaii International Conference on System Sciences. Honolulu, HI. Diamantopoulos, A. and Winklhofer, H.M. (2001). Index Construction with Formative Indicators: An Alternative to Scale Development. Journal of Marketing Research, 38(2), 269-277. Dibbern, J., Gales, T., Hirschheim, R. and Jayatilaka, B. (2004). Information Systems Outsourcing: A Survey and Analysis of the Literature. The DATA BASE for Advances in Information Systems, 34(4), 6-102. Edwards, J.R. (2001). Multidimensional Constructs in Organizational Behavior Research: An Integrative Analytical Framework. Organizational Research Methods, 4(2), 144-192. Feeney, D. and Willcocks, L.P. (1998). Core IS Capabilities for Exploiting Information Technology. Sloan Management Review, 1, 9-21 Ghosh, B. and Scott, J.E. (2004). Business Process Outsourcing -An Organizational Boon or Bust? INFORMS Annual Meeting (210). Denver, CO. Ghosh, B. and Scott, J.E. (2005a). Evaluating Business Process Outsourcing using Coordination Theory, Proceedings of the Eleventh Americas Conference on Information Systems. Omaha, NE. Ghosh, B. and Scott, J.E. (2005b). Comparing Knowledge Management in Healthcare and Technical Support Organizations. IEEE Transactions on Information Technology in BioMedicine, 9(2), 162-168. Ghosh, B. and Scott, J.E. (2005c). Effective Knowledge Management for Hospital Nurses. Proceedings of the Eleventh Americas Conference on Information Systems, Omaha, NE. 98

PAGE 110

Ghosh, B. and J.E. Scott (2005d). Inter Organizational Knowledge Management in a BPO. Proceedings of the Eleventh Americas Conference on Information Systems, Omaha, N E. Ghosh, B. and Scott, J.E.(2006a). An Exploratory Study of Organizational Impacts on BPO using Coordination Theory. Proceedings of the Twelfth Americas Conference on Information Systems, Acapulco, Mexico. Ghosh, B. and Scott, J.E. (2006b). Managing Clinical Knowledge Among Hospital Nurses. International Journal of Information Technology Management, accepted for publication. Ghosh, B. and Scott, J.E. (2007a). Effective Knowledge Management Systems for Nurses. Information Systems Management, 24, 73-84. Ghosh, B. and Scott, J.E. (2007b). Outsourcing. Proceedings of Systems, Keystone, Colorado. Social Capital in Knowledge based Business Process the Thirteeenth Americas Conference on Information Ghoshal, S., Korine, H., Szulanski, G. (1994). lnterunit Communication in Multinational Corporations, Management Science, 40(1), 96-110. Ghoshal, S. and Moran, P. (1996. Bad for Practice: A Critique of the Transaction Cost Theory. The Academy of Management Review, 21 (1), 13-47. Gold, A.H., Malhotra, A. and Segars, A.H. (2001). Knowledge management: an organizational capabilities perspective. Journal of Management Information Systems, 18(1 ), 185-214. Grant, R. M. (1991). The Resource-Based Theory of Competitive Advantage. California Management Review, 33(3), 114-135. Grover, V. and Davenport, T.H. (2001). General perspectives on knowledge management: Fostering a research agenda. Journal of Management Information Systems, 18(1), 5-22. Grover, V., Cheon, M. J. and Teng, J.T.C. (1996). The effect of Service Quality and Partnership on the Outsourcing of Information Systems Functions. Journal of Management Information Systems, 12(4), 89-116. Gulati, R., Nohria, N. and Zaheer, A. (2000). Strategic Networks. Strategic Management Journal (21 ), 203-215. Heckman, R. and King, W. (1994). Behavioral Indicators of Customer Satisfaction with Vendor Provided Information Services. Proceedings of the 15th International Conference on Information Systems (429-444). Vancouver, Canada. Hislop, D. (2002). Mission Impossible? Communicating and Management Knowledge via Information technology" Journal of Information Technology (17), 165-177. 99

PAGE 111

Holsapple, C.W. and Joshi, K.D. (2000). An investigation of factors that influence the management of knowledge in organizations. Journal of Strategic Information Systems, 9(2-3), 235-261. Hu, Q., Saunders, C. and Gebelt, M. (1997). Research Report: Diffusion of Information Systems Outsourcing: A Reevaluation of Influence Sources. Information Systems Research, 8(3), 288-301. Huber, G.P. (1991 ). Organizational learning: the contributing processes and the literature. Organization Science, 2(1), 88-115. Huber, R. L. (1993). How Continental Bank Outsourced its Crown Jewels. Harvard Business Review, 71(1), 121-129. Huang, S. and DeSanctis, G. (2005). Mobilizing Informational Social Capital in Cyber Space: Online Social Network Structural Properties and Knowledge Management. Proceedings of the Twenty-Sixth International Conference on Information Systems. Las Vegas, NV. Huysman, M. and Wulf, V. (2006). IT to support knowledge management in communities, towards a social capital analysis. Journal of Information Technology, 21, 40-51. lnkpen, A.C. and Tsang, E.W.K. (2005). Social Capital, Networks and Knowledge Transfer. Academy of Management Review, 30(1), 146-165. Jarvenpaa, S.L. and Leidner, D.E. (1999). Communication and Trust in Global Virtual Teams. Organization Science, 10(6), 791-815. Jarvenpaa, S.L. and Staples, D.S. (2001). Exploring perceptions of organizational ownership of information and expertise. Journal of Management Information Systems, 18(1), 151-184. Jarvenpaa, S.L. & Staples, D.S. (2000). The Use of Collaborative Electronic Media for Information Management: An Exploratory Study of Determinants. Journal of Strategic Information Systems, 9(2-3), 129-154. Jarvis, G.B., Mackenzie, S.B., Podsakoff, P.M. (2003). A critical review of Construct Indicators and Measurement Model Misspecification in Marketing and Consumer Research. Journal of Consumer Research, 30(2), 199-218. Kanhanhali, A., Tan, B.C.Y. and Wei, K.K. (2005). Contributing Knowledge to Electronic Knowledge Repositories: An Empirical Investigation. MIS Quarterly, 29(1), 113-143. Karimi, J., Somers, T.M., and Bhattacharjee, A. (2007a). The Role of Information Systems Resources in ERP Capability Building and Business Process Outcomes. Journal of Management Information Systems, forthcoming. Karimi, J., Somers, T.M., and Bhattacharjee, A. (2007b). The Impact of ERP Implementations on Business Process Outcomes: A Factor-Based Study. Journal of Management Information Systems, 24(1), 1 03-136. 100

PAGE 112

Kern, T. (1997). The Gestalt of an Information Technology Outsourcing Relationship: an Exploratory Analysis. Proceedings of the 1ih International Conference on Information Systems. Atlanta, Georgia. Kern, T. and Willcocks, L.P. (2000). Exploring Information Technology Outsourcing relationships: Theory and practice. Journal of Strategic Information Systems, 9(4), 321-350. Ko, D.G., Kirsch, L.J. and King, W.R. (2005). Antecedents of Knowledge Transfer from Consultants to Clients in Enterprise System Implementations. MIS Quarterly, 29(1), 59-85. Koh, C., Ang, S. and Straub, D.W. (2004). IT Outsourcing Success: A Psychological Contract Perspective. Information Systems Research, 15(4), 356-373. Kowch, E.G. (2005). Do we plan the journey or read the compass? An argument for preparing educational technologists to lead organizational change. Research Journal of Educational Technology, 36(6), 1067-1070. Kumar, K., van Dissel, H. G., and Bielli, P. (1998). The Merchant of Prato-Revisited: Toward a Third Rationality of Information Systems. MIS Quarterly, 22(2), 199-226. Lacity, M.C. and Willcocks, L.P. (1995). IT Outsourcing: Maximize Flexibility and Control. Harvard Business Review, 73(3), 84-93. Lacity, M.C. and Willcocks, L.P. (1998). An Empirical Investigation of Information Technology Sourcing Practices: Lessons form Experience. MIS Quarterly, 22(3), 363-408. Lave, J. and Wenger, E.(1991). Situated Learning: Legitimate Peripheral Participation. Cambridge, UK: Cambridge University Press. Lavie, D. (2006). The competitive Advantage of interconnected firms: An extension of the Resource Based View. Academy of Management Review, 31(3), 638-658. Lee, A. and Baskerville, R.L. (2003). Generalizing Generalizability in Information Systems Research. Information Systems Research, 14(3), 221-243. Lee, J. Nand Kim, Y. G. (1999). Effect of Partnership Quality on IS Outsourcing Success. Journal of Management Information Systems, 15(4), 29-61. Levina, N. and Ross, J.W. (2003). From the Vendor's Perspective: Exploring the Value Proposition in Information Technology Outsourcing. MIS Quarterly, 27(3), 331-364. Lieberman, J. I. (2004). Offshore Outsourcing and America's Competitive Edge: Losing out in the High Technology R&D and Services Sectors. Retrieved March 11, 2007 from US Senate White Paper, http://lieberman .senate. gov/documents/wh itepapers/Offshoring. pdf. Lin, H. (2006). lnterorganizational Collaboration, Social Embeddedness and Value Creation: A Theoretical Analysis. International Journal of Management, 23(3), 548-558. 101

PAGE 113

Loh, L. and Venkatraman, N. (1992). Determinants of Information Technology Outsourcing. Journal of Management Information Systems, 9(1), 7-24. Lyytinen, K., L. Mathiassen and J. Ropponen (1998). Attention Shaping and Software Risk -A Categorical Analysis of Four Classical Risk Management Approaches. Information Systems Research, (9)3, 233-255. Malhotra, A., Majchrzak, A., Carman, R. and Lott, V. (2001). Radical Innovation without Collocation: A Case Study at Boeing-Rocketdyne. MIS Quarterly 25(2), 229-249. Malhotra, A., Gosain, S. and El Sawy, O.A. (2005). Absorptive Capacity Configurations in Supply Chains: Gearing for Partner-Enabled Market Knowledge Creation. MIS Quarterly, 29(1), 145-187. Malone, T.W. (1987). Modeling Coordination in Organizations and Markets. Management Science, 33(10), 1317-1332. Malone, T. W., K. Crowston. (1994). The interdisciplinary study of coordination. ACM Computing Surveys, 26 (1), 87-119. Malone, T. W., K. Crowston. (1999). Tools for Inventing Organizations: Toward a Handbook of Organizational Processes. Management Science, 45(3), 425-444. Melville, N., Kraemer, K., and Gurbaxani, V. (2004). Review: Information Technology and Organizational Performance: An Integrative Model of IT Business Value. MIS Quarterly, 28(2), 283-322. Moore, S. (2004). IT Trends 2004: Offshore Outsourcing. Giga Research. Moore, S. and Martorelli, W. (2004). Indian Offshore Suppliers: The Market Leaders. FORRESTER Research Report. Nahapiet, J. and Ghoshal, S. (1998). Social Capital, Intellectual Capital and the Organizational Advantage. Academy of Management Review, 23(2), 242-268. Nidumolu, S.R., Subramani, M. and Aldrich, A. (2001). Situated learning and the situated knowledge web: exploring the ground beneath knowledge management. Journal of Management Information Systems, 18(1), 115-150. Nonaka, I. (1994). A dynamic theory of organizational knowledge creation. Organization Science, 5(1), 14-37. O'Brien, D.J., Phillips, J.L. and Patsiorkovsky, V.V. (2005). Linking Indigenous Bonding and Bridging Social Capital. Regional Studies, 39(8), 1041-1051. Orton, J.D. and Weick, K.E. (1990). Loosely Coupled Systems: A Reconceptualization. Academy of Management Review, 15(2), 203-223. 102

PAGE 114

Putnam, R. D. (1993). The prosperous community: Social capital and public life. The American Prospect, 4(13). Ray, G., T. Muhanna, W., and Barney, J. B. (2005). Information Technology and the Performance of the Customer Service Process: A Resource-Based Analysis. MIS Quarterly, 29(4), 625651. Ross, J. W.; Beath, C. M.; and Goodhue, D. L. (1996). Developing Long-term Competitiveness through IT Resources. Sloan Management Review, 38(1), 31-45. Rottman, J.W. and Lacity, M.C. (2004). Twenty Practices for Offshore Sourcing. MIS Quarterly Executive 3(3), 117-130. Sabherwal, R. (1999). The Role of Trust in Outsourced IS Development Project Communications of the ACM, 42(2), 80-86. Sambamurthy, V., Bharadwaj, A. and Grover, V. (2003). Shaping Agility Through Digital Options: Reconceptualizing the Role of Information Technology in Contemporary Firms. MIS Quarterly, 27(2), 237-263. Schmid, A. A. (2000). Affinity as social capital: its role in development The Journal of Socio Economics (29), 159. Schultze, U. and Boland, R.J. Jr. (2000). Knowledge management technology and the reproduction of knowledge work practices. Journal of Strategic Information Systems, 9(2-3), 193-212. Scott, J. ( 1998). Organizational Knowledge and the Intranet. Decision Support Systems, 23, 3-17. Scott, J.E. and Ghosh, B. (2004). Two dimensions of Knowledge Management: Technology and Organizational Culture. INFORMS Annual Meeting (180). Denver, CO. Segal, R., Bingham, B., Mayo, S. Danilenko, A., Romala, R., Motsenigos, A. and Tosto, S. (2003). Customer Demand for Solution and Business Process Outsourcing. /DC document # 29464. Snider, K.F. and Nissen, M.E. (2003). Beyond the body of knowledge: a knowledge-flow approach to project management theory and practice. Project Management Journal 34(2), 4-13. Szulanski, G. (2000). The Process of Knowledge Transfer: A Diachronic Analysis of Stickiness. Organizational Behavior and Human Decision Processes, 82(1), 9-27. Szulanski, G. (1996). Exploring Internal Stickiness: Impediments to the transfer of best practices within the firm. Strategic Management Journal, 17, 27-43 Tanriverdi, H. (2005). Information Technology Relatedness, Knowledge Management Capability, and Performance of Multi business Firms. MIS Quarterly, 29(2), 331-334. 103

PAGE 115

Tanriverdi, H., and Venkatraman, N. (2005), Knowledge Relatedness and Periormance of Multibusiness Firms. Strategic Management Journal (26), 97-119. Teo, H., Wei, K. and Benbasat, I. (2003). Predicting Intention to Adopt lnterorganizational Linkages: An Institutional Perspective. MIS Quarterly 27(1 ), 19-49. Trapper, D. (2003). Worldwide and US IT Outsourcing Services 2004-2008 Forecast: A Potential Periect Storm. IDC, document# 28975. Tushman, M. and Nadler, D. (1978). Information Processing as an Integrating Concept in Organizational Design. Academy of Management Review, 3(3), 613-624. Wade, M. and Hulland, J. (2004). Review: The Resource Based View and Information Systems Research: review, Extension and Suggestions for Future Research. MIS Quarterly, 28(1), 107-142. Wasko, M.M. and Faraj, S. (2000). It is what one does: why people participate and help others in electronic communities of practice. Journal of Strategic Information Systems, 9(2-3), 155173. Wasko, M.M. and Faraj, S. (2005). Why Should I Share? Examining Social Capital and Knowledge Contribution in Electronic Networks of Practice. MIS Quarterly, 29(1), 35-57. Watson, S. and Hewett, K. (2006). A Multi-theoretical Model of Knowledge Transfer in Organizations: Determinants of Knowledge contribution and Knowledge Reuse. Journal of Management Studies 43(2), 142-173. Wenger, E. (1998). Communities of Practice. Cambridge, UK: Cambridge University Press. Willcocks, L.P., Hindle, J., Feeny, D. and Lacity, M.C. (2004). IT and Business Process Outsourcing: The Knowledge Potential. Information Systems Management, 21(3), 7-15. Willcocks, L.P., Lacity, M.C. and Kern, T. (1999). Risk mitigation in IT outsourcing strategy revisited: longitudinal case research at LISA. Journal of Strategic Information Systems, 8(3), 285314. Willcocks, L.P. and Kern, T. (1998). IT Outsourcing as Strategic Partnering: The Case of the UK Inland Revenue. European Journal of Information Systems, 5(1), 29-45. Yuan, Y. and McKelvey, B. (2002). Situated Learning Theory: Adding rate and Complexity Effects via Kauffman's NK Model. Nonlinear Dynamics, Psychology and Life Sciences, 12, 67-77. Zahra, S. A.; George, G. (2002). The Net-Enabled Business Innovation Cycle and the Evolution of Dynamic Capabilities. Information Systems Research, 13(2), 14 7-151 104

PAGE 116

Zhu, K. (2004). The Complementarity of Information Technology Infrastructure and E-Commerce Capability: A Resource-Based Assessment of Their Business Value. Journal of Management Information Systems, 21 (1), 167-202. 105