Kankanhalli ContributingKnowledgeElectronic 2005

Contributing Knowledge to Electronic Knowledge Repositories: An Empirical
Investigation
Author(s): Atreyi Kankanhalli, Bernard C. Y. Tan and Kwok-Kee Wei
Source: MIS Quarterly , Mar., 2005, Vol. 29, No. 1, Special Issue on Information
Technologies and Knowledge Management (Mar., 2005), pp. 113-143
Published by: Management Information Systems Research Center, University of
Minnesota
Stable URL: https://www.jstor.org/stable/25148670
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Management Information Systems Research Center, University of Minnesota is collaborating

with JSTOR to digitize, preserve and extend access to MIS Quarterly
This content downloaded from

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
All use subject to https://about.jstor.org/terms
Kankanhalli et al./Contributing to Electronic Knowledge Repositories
v Hid,! CwIlV secalssuE
Contributing Knowledge to Electronic

Knowledge Repositories: An Empirical
Investigation1
By: Atreyi Kankanhalli Abstract

Department of Information Systems
National University of Singapore Organizations are attempting to leverage their
3 Science Drive 2 knowledge resources by employing knowledge
Singapore 117543 management (KM) systems, a key form of which
REPUBLIC OF SINGAPORE are electronic knowledge repositories (EKRs). A
atreyi@comp.nus.edu.sg large number of KM initiatives fail due to the
reluctance of employees to share knowledge
Bernard C. Y. Tan through these systems. Motivated by such con
Department of Information Systems cerns, this study formulates and tests a theoretical
National University of Singapore model to explain EKR usage by knowledge
3 Science Drive 2 contributors. The model employs social exchange
Singapore 117543 theory to identify cost and benefit factors affecting
REPUBLIC OF SINGAPORE EKR usage, and social capital theory to account
btan@comp.nus.edu.sg for the moderating influence of contextual factors.
The model is validated through a large-scale
Kwok-Kee Wei survey of public sector organizations. The results
Department of Information Systems reveal that knowledge self-efficacy and enjoyment
City University of Hong Kong in helping others significantly impact EKR usage
83 Tat Chee Avenue by knowledge contributors. Contextual factors
Kowloon (generalized trust, pro-sharing norms, and identifi
HONG KONG SAR cation) moderate the impact of codification effort,
isweikk@cityu.edu.hk reciprocity, and organizational reward on EKR
usage, respectively. It can be seen that extrinsic
benefits (reciprocity and organizational reward)
impact EKR usage contingent on particular con
textual factors whereas the effects of intrinsic
benefits (knowledge self-efficacy and enjoyment in
helping others) on EKR usage are not moderated
1V. Sambamurthy and Mani Subramani were the

by contextual factors. The loss of knowledge
accepting senior editors for this paper. power and image do not appear to impact EKR
MIS Quarterly Vol. 29 No. 1, pp. 113-143/March 2005 113

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
usage by knowledge contributors. Besides contri facilitate knowledge reuse through access to the
buting to theory building in KM, the results of this codified expertise. A key technological component
study inform KM practice. of this approach is electronic knowledge reposi
tories (EKRs) (Grover and Davenport 2001). The
Keywords: Knowledge management, electronic network model corresponds to the personalization
knowledge repositories, knowledge contribution, approach to KM (Hansen et al. 1999). This
social exchange, social capital approach emphasizes linkage among people for
the purpose of knowledge exchange. Important
technological components of this approach are
knowledge directories that provide location of
Introduction expertise (Ruggles 1998) and electronic forum
software that allows people to interact within
The strategic management of organizational communities of practice (Brown and Duguid 1991).
knowledge is a key factor that can help organiza
tions to sustain competitive advantage in volatile While technological capabilities are important,
environments. Organizations are turning to knowl having sophisticated KM systems does not
edge management (KM) initiatives and techno guarantee success in KM initiatives (Cross and
logies to leverage their knowledge resources. Baird 2000; McDermott 1999). This is because
According to the analyst firm IDC, business social issues appear to be significant in ensuring
spending on KM could rise from $2.7 billion in knowledge sharing success (Ruppel and Harring
2002 to $4.8 billion in 2007 (Babcock 2004). ton 2001). Both social and technical barriers to
Concurrent with the organizational interest in KM, usage of KM systems have been listed and calls
a large number of academic papers have been have been made to simultaneously address both
published on KM (Schultze and Leidner 2002). sets of issues (McDermott 1999; Zack 1999) in
These developments reflect the significance of KM order to be able to reap the benefits of KM that
among scholars and practitioners. have been experienced by some organizations
(Davenportetal. 1998; O'Dell and Grayson 1998).
Knowledge management is defined as "a systemic
and organizationally specified process for ac This study focuses on EKRs since they are
quiring, organizing, and communicating both tacit fundamental to organizational knowledge capture
and explicit knowledge of employees so that other and dissemination, yet the factors affecting EKR
employees may make use of it to be more effective usage are not well understood (Markus 2001).
and productive in their work" (Alavi and Leidner EKRs are electronic stores of content acquired
1999, p. 6). KM systems are "a class of informa about all subjects for which the organization has
tion systems applied to managing organizational decided to maintain knowledge (Liebowitz and
knowledge. That is, they are IT-based systems Beckman 1998). EKRs can comprise multiple
developed to support and enhance the organiza knowledge bases as well as the mechanisms for
tional processes of knowledge creation, storage/ acquisition, control, and publication of the knowl
retrieval, transfer, and application" (Alavi and edge.3 The process of knowledge sharing through
Leidner 2001, p. 114). Two models of KM systems EKRs involves people contributing knowledge to
have been identified in the information systems populate EKRs (e.g., customer and supplier
literature: the repository model and the network knowledge, industry best practices, and product
model (Alavi 2000).2 The repository model expertise) and people seeking knowledge from
corresponds to the codification approach to KM
(Hansen et al. 1999). This approach emphasizes
codification and storage of knowledge so as to 3According to the definition, the capabilities of EKRs are
analogous to the mnemonic functions of organizational
memory information systems (Stein and Zwass 1995).
However, conceptualization of organizational memory
2Zack (1999) alternately labels these two models as information systems concentrates more on description at
integrative and interactive architectures respectively. the subsystem level.
114 MIS Quarterly Vol. 29 No. 1/March 2005

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
EKRs for reuse. Success of EKRs requires that edge contributors. In investigating the usage of
knowledge contributors be willing to part with their EKRs, the first choice of theoretical bases would
knowledge and knowledge seekers be willing to appear to be theories such as the technology
reuse the codified knowledge (Ba et al. 2001). acceptance model (Davis 1989) which have been
The distinction between contributors and seekers successful in explaining the usage of information
is conceptual in that the same individual can be a systems (e.g., Venkatesh and Davis 2000).
contributor or a seeker at different points in time. Although the technology acceptance model may
This study examines EKR usage from the partially explain the behavior of knowledge
perspective of knowledge contributors as this is contributors,4 this model does not directly account
the first step toward knowledge leverage through for the social cost and benefit factors experienced
EKRs. Unless knowledge contributors are willing by knowledge contributors that may affect their
to provide content to EKRs, knowledge reuse usage of collective technologies such as EKRs.
through EKRs cannot take place. However, the social and individual cost and benefit
factors in knowledge sharing can be accounted for
Several prior studies have adopted a conceptual by social exchange theory. The impact of cost and
(Kollock 1999; Markus 2001) or qualitative ap benefit factors on EKR usage by knowledge
proach (Goodman and Darr 1998; Orlikowski contributors is likely to be contingent upon
1993; Wasko and Faraj 2000) in attempts to contextual factors (Constant et al. 1996; Goodman
understand the behavior of knowledge contribu and Darr 1998; Jarvenpaa and Staples 2000;
tors. Other studies have conducted experiments Orlikowski 1993). Social capital theory accounts
(Constant et al. 1994) or surveys (Bock et al. 2005; for several important contextual factors in
Constant et al. 1996; Jarvenpaa and Staples 2000; knowledge exchange. Therefore, this study uses
Wasko and Faraj 2005) to model and explain the social exchange theory and the social capital
contributor behavior with varying success. theory as its theoretical bases.
Existing empirical studies focus on the benefits
(acting as motivators) rather than the costs (acting
as inhibitors) of knowledge contribution, despite
suggestions from practitioner literature (O'Dell and Cost and Benefit Factors
Grayson 1998) and conceptual literature (Ba et al.
2001) that cost factors can be important in Cost and benefit factors in our study are derived
determining knowledge-sharing behavior. This based on social exchange theory. Social ex
study advances theoretical development on change theory explains human behavior in social
knowledge contribution in two important ways. exchanges (Blau 1964), which differ from eco
First, it simultaneously investigates both cost and nomic exchanges in that obligations are not clearly
benefit factors affecting EKR usage. Second, it specified. In such exchanges, people do others a
incorporates contextual factors to illustrate how favor with a general expectation of some future
these may moderate the relationships between return but no clear expectation of exact future
cost and benefit factors and EKR usage. The return. Therefore, social exchange assumes the
results suggest organizational interventions and existence of relatively long-term relationships of
technology design considerations that can promote interest as opposed to one-off exchanges (Molm
knowledge contribution to EKRs, thereby faci 1997). Knowledge sharing through EKRs can be
litating reuse of organizational knowledge. seen as a form of generalized social exchange
(Fulk et al 1996) where more than two people
participate and reciprocal dependence is indirect,
Theoretical Background 1HHH

4The technology acceptance model has been suggested
The dependent variable we are interested in
as a means to extend the theoretical model developed in
investigating is the degree of EKR usage by knowl this study.
MIS Quarterly Vol. 29 No. 1/March 2005 115

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
with the EKR serving as the intermediary between rewards for their contributions (Beer and Nohria
knowledge contributors and seekers. Knowledge 2000; Hall 2001) through which they can obtain a
contributors share their knowledge with no exact better lifestyle. As a result of contribution, knowl
expectation of future return. The quantity and edge contributors may also enhance their image or
value of knowledge contributed is difficult to reputation in the organization (Ba et al. 2001;
specify and so is the return obtained. Hence, Constant et al. 1994; Constant et al. 1996), which
knowledge contributors are likely to work on the can serve to increase their self-concept. By
assumption of relatively longer-term relationships sharing their knowledge, contributors may receive
of interest. reciprocal benefits, i.e., their future requests for
knowledge being met by others (Connolly and
Resources (tangible and intangible) are the cur Thorn 1990; Kollock 1999; Wasko and Faraj
rency of social exchange. Resources given away 2000), which can facilitate their work. Intrinsic
during social exchange or negative outcomes of benefits are sought after as ends by themselves.
exchange can be seen as costs. Resources For example, through contribution, knowledge con
received as a result of social exchange or positive tributors can be satisfied by enhancing their knowl
outcomes of exchange can be seen as benefits. edge self-efficacy or confidence in their ability to
Social exchange theory posits that people behave provide valuable knowledge that is useful to the
in ways that maximize their benefits and minimize organization (Constant et al. 1994; Constant et al.
their costs (Molm 1997). In agreement with this 1996). Also, by contributing knowledge to EKRs,
theory, researchers have suggested that in knowledge contributors have the opportunity to
creasing the benefits and reducing the costs for help others (Ba et al. 2001; Wasko and Faraj
contributing knowledge can help to encourage 2000). Previous studies on altruism have shown
knowledge sharing using KM systems (Markus that people enjoy and derive pleasure from such
2001; Wasko and Faraj 2000), including EKRs. acts of helping others (Baumeister 1982; Krebs
1975). Research has established extrinsic and
During social exchange, costs can be incurred in intrinsic benefits as motivators of human behavior
the form of opportunity costs and actual loss of in several domains (Vallerand 1997), including
resources (Molm 1997). Opportunity costs are knowledge sharing (Osterloh and Frey 2000).
rewards foregone from alternative behavior not
chosen. For example, the time and effort required
to codify and input knowledge into EKRs (Ba et al.
2001; Markus 2001) can act as an opportunity cost Contextual Factors
that precludes knowledge contributors from per
forming alternative tasks at that time and accruing Contextual factors in our study are derived from
the corresponding rewards. Also, knowledge con social capital theory. Social capital refers to the
tributors may perceive a loss of power and unique resources embedded within networks of human
value within the organization associated with the relationships (Nahapiet and Ghoshal 1998).
knowledge they transfer to EKRs (Davenport and These networks include proximate as well as
Prusak 1998; Gray 2001). Such loss of knowledge virtual communities (Rheingold 2000). Social
power can be considered as an actual loss of capital theory posits that social capital provides the
resource during knowledge contribution. conditions necessary for knowledge exchange to
occur. Three key aspects of social capital that can
During social exchange, benefits acting as moti define the context for knowledge exchange are
vators of human behavior can be extrinsic or trust, norms, and identification (Nahapiet and
intrinsic in nature (Deci and Ryan 1980; Vallerand Ghoshal 1998). Trust, norms, and identification
1997). Extrinsic benefits are sought after as can be considered as social capital since they are
means to ends desired by people. For example, organizational resources or assets rooted within
knowledge contributors may receive organizational social relationships that can improve the efficiency

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
of coordinated action. Practitioner literature has A norm represents a degree of consensus in the
described the impacts of these factors without social system (Coleman 1990). Norms have the
considering whether their effects are direct or effect of moderating human behavior in accor
moderating. However, several prior academic dance with the expectations of the group or
studies (e.g., Constant et al. 1994; Jarvenpaa and community. Pro-sharing norms that have been
Staples 2000) have hinted at the moderating role reported to enhance the climate for knowledge
of these aspects of social capital in knowledge sharing are norms of teamwork (Starbuck 1992),
sharing situations. Specifically, these three factors collaboration and sharing (Goodman and Darr
are believed to be able to amplify or dampen the 1998; Jarvenpaa and Staples 2000; Orlikowski
effects of particular cost and benefit factors on 1993), willingness to value and respond to
knowledge-sharing behavior. diversity, openness to conflicting views, and
tolerance for failure (Leonard-Barton 1995). The
Trust is the belief that the intended action of others effects of costs and extrinsic benefits on knowl
would be appropriate from our point of view edge contribution behavior are likely to be con
(Mistzal 1996). It indicates a willingness of people tingent on these norms. The effects of intrinsic
to be vulnerable to others due to beliefs in their benefits on knowledge contribution are not likely to
good intent and concern, competence and capa be affected by contextual factors since these
bility, and reliability (Mishra 1996). McKnight etal. benefits are seen as ends in themselves. The cost
(1998) term these trusting beliefs as benevolence and extrinsic benefit factors may not influence
belief, competence belief, and predictability belief, contribution behavior under conditions of strong
respectively. Generalized trust is an impersonal pro-sharing norms. Specifically, when pro-sharing
form of trust that does not rest with a specific norms are strong, the costs of knowledge sharing
individual but rests on behavior that is generalized may not be a deterrent to knowledge contributors.
to a social unit as a whole (e.g., a community of Conversely when pro-sharing norms are weak, the
knowledge workers exchanging knowledge costs of contribution may be salient. For example,
through EKRs) (Putnam 1993). In the context of it has been found that the costs of knowledge
our study, generalized trust refers to the belief in sharing were a deterrent to KM system usage
the good intent, competence, and reliability of when there was incompatibility between the
employees with respect to contributing and reusing collective nature of the technology and the com
knowledge through EKRs. With strong gener petitive norm of the organizational context
alized trust, people may trust each other without (Orlikowski 1993). The extrinsic benefits of knowl
having much personal knowledge about each edge sharing may not be salient to knowledge
other. Generalized trust has been viewed as a key contributors when pro-sharing norms are strong.
factor that provides a context for cooperation (Tsai For example, in environments where such norms
and Ghoshal 1998) and effective knowledge are strong, there is greater openness to the
exchange (Adler 2001). When generalized trust is potential for value creation through knowledge
strong, the effort required for knowledge sharing exchange (Nahapiet and Ghoshal 1998) even
may not be salient to knowledge contributors when extrinsic benefits to stimulate knowledge
because they believe that knowledge shared is not contribution may not exist.
likely to be misused by reusers (Davenport and
Prusak 1998). Conversely, when generalized trust Identification is a condition where the interests of
is weak, knowledge contributors may find the effort individuals merge with the interests of the
required for knowledge sharing to be salient organization, resulting in the creation of an identity
because they believe that others may inappro based on those interests (Johnson et al. 1999).
priately use their knowledge. For example, con Identification sets the context within which com
sultants at Ernst and Young declined to make the munication and knowledge exchange occur among
effort to contribute knowledge to repositories in organizational members (Nahapiet and Ghoshal
situations where trust did not exist (Markus 2001). 1998). Three components of identification that

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
have been identified in the literature are similarity contributors contingent on particular contextual
of values, membership in the organization, and factors (i.e., generalized trust, pro-sharing norms,
loyalty toward the organization (Patchen 1970). and identification).
Similarity of values reflects the extent to which
members of an organization possess joint goals
and interests. Membership is the degree to which
self-concept of members is linked to the organiza Loss of Knowledge Power
tion. Loyalty refers to the extent to which mem
Previous research suggests that by contributing a
bers support and defend the organization.
part of their unique knowledge, knowledge con
Identification is likely to provide a context for pro
tributors give up sole claim to the benefits
social behavior by raising the concern for collective
stemming from such knowledge (Gray 2001).
interests which merge with the individual's own
Therefore, knowledge contributors retain less
interests (Johnson et al. 1999; O'Reilly and Chat
proprietary knowledge upon which to argue their
man 1986). Under conditions of strong identifica
value to the organization. This may reduce the
tion, the effects of certain costs and benefits
power position of knowledge contributors in
pertaining to knowledge sharing may be nullified in
relation to the organization, making them more
the face of collective outcomes (Constant et al.
replaceable. These arguments have also been
1996). Therefore, when identification is strong, the
echoed by scholars in economics (Williamson
effort required for knowledge sharing may not be 1975) and organization strategy (Mintzberg 1973;
a deterrent to knowledge contributors because the Pfeffer 1992) in that uniqueness is considered a
concern for organizational outcomes may domi key aspect of organizational power because the
nate. Similarly, in such contexts, the need for lower the substitutability of an individual, the
organizational reward for knowledge sharing may greater is his or her power (Hickson et al. 1971).
not be salient to knowledge contributors because The KM literature reports the loss of power due to
the regard for collective outcomes is strong. knowledge contribution as a barrier to knowledge
sharing (Davenport and Prusak 1998; Orlikowski
1993). Since knowledge is perceived as a source
of power, knowledge contributors may fear losing
Research Model their power or value if others know what they know
(Gray 2001; Thibaut and Kelley 1986). Potential
and Hypotheses knowledge contributors may keep themselves out
of a knowledge exchange if they feel they can
The research model for explaining EKR usage by
benefit more by hoarding their knowledge rather
knowledge contributors incorporates constructs
than by sharing it (Davenport and Prusak 1998).
from social exchange theory and social capital
theory (see Figure 1). Previous studies have
While the above discussion suggests a negative
emphasized the importance of cost and benefit
relationship between loss of knowledge power and
factors in determining knowledge sharing behavior
EKR usage by knowledge contributors, this rela
(Ba et al. 2001; Markus 2001). Prior research has
tionship may be contingent on pro-sharing norms.
also highlighted the importance of contextual When such norms are strong, the barriers to
social capital factors in influencing the conditions
knowledge transfer witnessed in contexts that
for knowledge sharing (Cohen and Prusak 2001; value personal expertise may be ineffective
Nahapiet and Ghoshal 1998). Therefore, cost (Jarvenpaa and Staples 2000). In such conditions
factors (i.e., loss of knowledge power and codifi when other employees are seen to be sharing
cation effort) and benefit factors (i.e., organiza knowledge, the deterrent effect of the loss of
tional reward, image, reciprocity, knowledge self knowledge power (which is relative to other em
efficacy, and enjoyment in helping others) are ployees) may not be significant. Conversely with
hypothesized to impact EKR usage by knowledge weak pro-sharing norms, knowledge contributors

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Costs
Loss of Knowledge Power ^

Pro-sharing Norms "^^^^-^^^
I-:-1 ^""^ H1 l?I
Codification Effort ^^^^^ E
Generalized Trust ""^--^^^ ^^^^^ *
Codification Effort H2a -^^^ u
Pro-sharing Norms """"??-._ ?,^^ s
Codification Effort -_ E
H2b-___^^jj g
Identification H2c *
1-' B
I-1 Y
Extrin
N
I-1 ?
Organiz
Pro-sharing Norms H3a L
._. D
Identification
Org
1 ' c
| [ _^ o
Image* _-H4-" r n
Pro-sharing Norms T
I-1 R
Reciprocity _ - H5- B
Pro-sharing Norms ~" ^
Intrinsic Benefits ^-~ H6 _

Knowledge Self-efficacy - ^^. H7
Enjoyment in Helping ^-^"^
|_Others_|
Figure 1. The Research Model for EKR Usage by Knowledge Contr
MIS Quarterly Vol. 29 No. 1/March 20

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
may be concerned about loss of knowledge power H2a: Codification effort is negatively related to
when they contribute knowledge to EKRs. EKR usage by knowledge contributors under
conditions of weak generalized trust
HI: Loss of knowledge power is negatively
related to EKR usage by knowledge Codification effort may also be negatively related
contributors under conditions of weak pro to EKR usage by knowledge contributors contin
sharing norms. gent on pro-sharing norms. With strong teamwork
as well as cooperation and collaboration norms,
people tend not to be bothered about the effort
required to contribute knowledge since others may
Codification Effort
be likewise contributing. Conversely, when weak
pro-sharing norms prevail, knowledge contributors
The act of knowledge contribution to EKRs in
tend to be aware of and hindered by the effort
volves explication and codification of knowledge.
required to codify knowledge for EKRs (Orlikowski
This can entail costs to knowledge contributors as
1993). Therefore, knowledge contributors may be
an expense of time and effort (Ba et al. 2001;
deterred by codification effort when pro-sharing
Markus 2001). Effort has been observed to be a norms are weak.
significant predictor of technology adoption
(Agarwal 2000). The time required for codifying
H2b: Codification effort is negatively related to
knowledge can be considered as an opportunity
EKR usage by knowledge contributors under
cost. Orlikowski (1993) reported a situation where
conditions of weak pro-sharing norms.
consultants avoided knowledge contribution due to
high opportunity cost. They were unwilling to use
Further, codification effort may be negatively
the KM system as this would have required them
related to EKR usage by knowledge contributors
to incur non-chargeable hours or give up their
contingent on identification. When identification is
personal time. After contributing knowledge, there
strong, collective concerns may dominate over
may be additional requests for clarification and
certain costs and benefits since collective interests
assistance from knowledge recipients, which take
merge with the individual's own interests (Johnson
up more codification time from knowledge con
et al. 1999). In such situations, knowledge contri
tributors (Goodman and Darr 1998).
butors may not be concerned about their codifi
cation effort in the presence of the collective need
The above arguments suggest a negative relation
for contributing knowledge to EKRs. Conversely,
ship between codification effort and EKR usage by
when identification is weak, codification effort may
knowledge contributors, but this relationship is
become a deterrent to knowledge contribution.
likely to be contingent on generalized trust. Strong
generalized trust implies a general belief in the
H2c: Codification effort is negatively related to
good intent of others (Putnam 1993). When
EKR usage by knowledge contributors under
generalized trust is strong, knowledge contributors conditions of weak identification.
are likely to believe that knowledge recipients
would not misuse their knowledge and would give
them credit for their codification effort. Under such
circumstances, knowledge contributors may be Organizational Reward
confident that their effort would be appreciated and
may not be concerned about the effort they have To encourage EKR usage by knowledge contri
to put in when contributing knowledge to EKRs. butors, organizations may provide various forms of
Conversely, codification effort may become a organizational reward such as increased pay,
deterrent to knowledge contribution when gener bonuses, job security, or career advancement (Ba
alized trust is weak. et al. 2001; Beer and Nohria 2000; Hall 2001).

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
The American Productivity Quality Center website Image

provides examples of reward schemes for en
couraging knowledge sharing in organizations (see In most organizations today, the importance of
http://www.apqc.org/). Several consulting com reputation is increasing as traditional contracts
panies have made knowledge sharing a basic between organizations and employees based on
criterion for employee performance evaluation length of service erode (Ba et al. 2001; Davenport
(Davenport and Prusak 1998). There are also in et al. 1998). In such working environments, knowl
stances where employees enjoy better job security edge contributors can benefit from showing others
as a result of contributing knowledge (Hall 2001). that they possess valuable expertise (Ba et al.
2001). This earns them respect (Constant et al.
Although this discussion suggests a positive rela 1994) and a better image (Constant et al. 1996).
tionship between organizational reward and EKR Therefore, knowledge contributors can benefit
usage by knowledge contributors, the relationship from improved self-concept when they contribute
is likely to be contingent on pro-sharing norms. their knowledge (Hall 2001; Kollock 1999).
When norms of teamwork and collaboration are Employees have been found to share their best
strong, people may not require external incentives practice due to a desire to be recognized by their
to share knowledge (Nahapiet and Ghoshal 1998). peers as experts (O'Dell and Grayson 1998).
Such a climate can remove the need for organi People who provided high-quality technical knowl
zational reward because knowledge contributors edge have been found to enjoy better prestige in
may contribute their knowledge to EKRs even the workplace (Kollock 1999).
when such rewards are not available. Conversely,
weak pro-sharing norms can make organizational While this discussion suggests a positive relation
reward a salient motivator for knowledge ship between image and EKR usage by knowledge
contribution. contributors, the relationship may be contingent on
pro-sharing norms. When strong teamwork and
H3a: Organizational reward is positively related to collaboration norms prevail, knowledge contri
EKR usage by knowledge contributors under butors may not require extrinsic benefits (Nahapiet
conditions of weak pro-sharing norms. and Ghoshal 1998) such as image in order to
contribute knowledge. Under such conditions,
Further, organizational reward may be positively knowledge contributors are likely to contribute their
related to EKR usage by knowledge contributors knowledge to EKRs even if benefit in the form of
contingent on identification. When identification is improved image is absent. Conversely, weak pro
strong, the interests of employees and the sharing norms can make image a salient motivator
interests of the organization merge (Johnson et al. for knowledge contribution.
1999). In such situations, collective concerns tend
to dominate over certain costs and benefits. H4: Image is positively related to EKR usage by
Therefore, under conditions of strong identification, knowledge contributors under conditions of
knowledge contributors may not require organi weak pro-sharing norms.
zational reward to motivate them. Even when
organizational reward is absent, people may still
contribute knowledge to EKRs. Conversely, weak
Reciprocity
identification can make organizational reward a
salient motivator for knowledge contributors.
Reciprocity has been highlighted as a benefit for
individuals to engage in social exchange (Blau
H3b: Organizational reward is positively related to 1964). It can serve as a motivational mechanism
EKR usage by knowledge contributors under for people to contribute to discretionary databases
conditions of weak identification. (Connolly and Thorn 1990). Reciprocity can act as

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
a benefit for knowledge contributors because they 1999). Knowledge self-efficacy is typically mani
expect future help from others in lieu of their fested in the form of people believing that their
contributions5 (Connolly and Thorn 1990; Kollock knowledge can help to solve job-related problems
1999). Prior research suggests that people who (Constant et al. 1996), improve work efficiency (Ba
share knowledge in online communities believe in et al. 2001), or make a difference to their organi
reciprocity (Wasko and Faraj 2000). Further, zation (Kollock 1999; Wasko and Faraj 2000).
researchers have observed that people who Conversely, if people feel that they lack knowledge
regularly helped others in virtual communities that is useful to the organization, they may decline
seemed to receive help more quickly when they

from contributing knowledge to EKRs because
they believe that their contribution cannot make a
asked for it (Rheingold 2000).
positive impact for the organization. The effect of
knowledge self-efficacy on EKR usage is not likely
These arguments suggest a positive relationship
to be contingent on generalized trust, pro-sharing
between reciprocity and EKR usage by knowledge
norms, or identification.
contributors, but the relationship may be contin
gent on pro-sharing norms. When pro-sharing
H6: Knowledge self-efficacy is positively related
norms are strong, knowledge contributors may
to EKR usage by knowledge contributors.
share their knowledge without need for extrinsic
benefits (Nahapiet and Ghoshal 1998) such as
reciprocity. In such a climate, knowledge contri
butors are likely to contribute their knowledge to
Enjoyment in Helping Others
EKRs even in the absence of reciprocity benefits.
Conversely, when pro-sharing norms are weak, This benefit is derived from the concept of
reciprocity may be a salient motivator for knowl altruism. Altruism exists when people derive
edge contributors.
intrinsic enjoyment from helping others without
expecting anything in return (Krebs 1975; Smith
H5: Reciprocity is positively related to EKR
1981). Although there may be very few instances
usage by knowledge contributors under
of absolute altruism (involving absolute lack of self
conditions of weak pro-sharing norms.
concern in the motivation for an act), relative
altruism (where self-concern plays a minor role in
motivating an act) is more prevalent (Smith 1981).
Knowledge Self-Efficacy Knowledge contributors may be motivated by rela
tive altruism based on their desire to help others
Self-efficacy relates to the perception of people (Davenport and Prusak 1998). Prior research
about what they can do with the skills they possess shows that knowledge contributors gain satis
(Bandura 1986). When people share expertise faction by demonstrating their altruistic behavior
useful to the organization, they gain confidence in (Wasko and Faraj 2000). Such satisfaction stems
terms of what they can do and this brings the from their intrinsic enjoyment in helping others (Ba
benefit of increased self-efficacy (Constant et al. et al. 2001; Constant et al. 1994; Constant et al.
1994). This belief can serve as a self-motivational 1996). Knowledge contributors who derive enjoy
force for knowledge contributors to contribute ment in helping others may be more inclined to
knowledge to EKRs (Bock and Kim 2002; Kalman contribute knowledge to EKRs. The effect of
enjoyment in helping others on EKR usage is not
likely to be contingent on generalized trust, pro
Reciprocity refers to the expectation of knowledge sharing norms, or identification.
contributors that their current contribution will lead to
their future request for knowledge being met. This is
different from pro-sharing norms, which describe a HI: Enjoyment in helping others is positively
climate to facilitate knowledge sharing whereby sharing
related to EKR usage by knowledge
may take place with little consideration for the costs and
contributors.
extrinsic benefits (including reciprocity) of doing so.

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Table 1. Formal Definitions of Constructs

Construct
(Abbreviation) Definition
Loss of knowledge The perception of power and unique value lost
power (LOKP) to EKR (Gray 2001)
Codification effort The time and effort required to codify and inpu
(CEFF) 2001)
Organizational The importance of economic incentives provided for kn
reward (OREW) to EKR (Ba et al 2001; Hall 2001)
Image (IMAG) The perception of increase in reputation due to contrib
EKR (Constant et al 1996; Kollock 1999)
Reciprocity (RECP) The belief that current contribution to EKR would lead
knowledge being met (Davenport and Prusak 1998)
Knowledge self- The confidence in one's ability to provide knowledge t

efficacy (KSEF) organization via EKR (Constant et al. 1996; Kalman
Enjoyment in helping The perception of pleasure obtained from helping ot
others (EHLP) contributed to EKR (Wasko and Faraj 2000)
Generalized trust The belief in the good intent, competence, and reliability
(GTRU) respect to contributing and reusing knowledge (Mishra
Pro-sharing norms The prevalence of norms that are intended to facilitate
(PSNM) the organization (Nahapiet and Ghoshal 1998; Orlikow
Identification (IDEN) The perception of similarity of values, membership, an
organization (Johnson et al 1999; Patchen 1970)
EKR usage (EUSG) The degree of EKR usage to contribute knowledge
Research Methodologywhere, new questions were developed ba

review of the previous KM and info
The survey methodology was used to systems
collect data
literature. One question for
for testing the research hypotheses. This method
by knowledge contributors (the depen
ology was chosen because itstruct) enhances
measured the frequency of EK
generalizability of results (Dooley 2001).
The remaining 50 questions in the instru
measured using seven-point scales ancho
"strongly disagree" to 'strongly agree
appendix).
Operationalization of Constructs
Table 1 provides formal definitions of the con

structs. Where available, these constructs were
measured using questions adapted from prior
6We have to be mindful of the limitation that self-reported
studies to enhance validity (StoneEKR
1978).
usage mayElse
be inflated.

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Table 2. Results of Unstructured Sorting Exercise

Actual Category
0-l_5_)CLl_,Q.3S-pOjK Hit
Target __i_iyôiii-jo_zgc02 Total Rate
Category 3 S O I S S 5 5 2 Q 5 5 Qs
LOKP 12 2 2 16 75
CEFF 15 1 4 20 75
OREW 20 20 100
IMAG 2 14 1 3 20 70
RECP 6 2 8 75
KSEF 14 1 1 16 87.50
EHLP 16 16 100
GTRU 16 16 100
PSNM 1 20 3 24 83.33
IDEN 1 31 4 36 86.11
EUSG 12 12 100
Average 86.54
Conceptual Validatio
IDEN5) were dr
these questions
Given that the
withquestions
the other
constructs were adapted from
developed for Another
this four
study, all g
subjected to a two-stage
second stage co
(s
exercise sorter
based on was g
procedu
Moore and printed (1991)
Benbasat on car
dents vious
participated in stage,
the t
firs
sorting) as definitions
sorters. Each of
sortt
questions questions
printed on by
cards
had the to
questions
construct sort
by
categ
tions together Apart
and from
giving t
a
related questions
IMAG4)(which
that m w
This process allhelped to corr
sorters iden
worded questions.
the The
intendedlabel
c
sorters for questions
the constructswer
closely to the names
from theof th
sorte
Overall, the procity
four sorters (RECP
corre
86 percent ofonthe question
suggestions
constructs desirable
(see Table to On
2). ha
image (IMAG5) per
and construct
pro-shari
and two additional
questions for ques
identi
124 MIS Quarterly Vol. 29 No

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Table 3. Results of Structured Sorting Exercise

Actual Category
Q-lL.3ti)Q-li_(L=>5.-e>5K Hlt
Target û-wSow-ifczgcoS Total Rate
Category 3 g ? I S S 5 O ? 8 5 5
LOKP 16 16 100
CEFF 20 20 100
OREW 20 20 100
IMAG 14 2 16 87.5
RECP 8 8 100
KSEF 16 16 100
EHLP 16 16 100
GTRU 16 16 100
PSNM 20 20 100
IDEN 28 28 100
EUSG 12 12 100
Average 98.86
ment from each organization

properties for helped to identify and
reciproc
were consolidated then into
distribute the survey to colleagues who had used
survey administration.
EKRs as knowledge contributors in the course of
their work. Hence, all respondents to this survey
were KM practitioners familiar with EKR usage
from the perspective of knowledge contribution.
Survey Administratio
Among the 400 surveys distributed in these
organizations, 150 responses were obtained
The study yielding a was
field response rate ofconducte
37.5 percent (see
a period of 6 Table months,
4). Completed surveys were returned tofrom
the
Singapore is a authors by
developedthe respondents using envelopes with
coun
of public pre-paid postage.
organizations are
embarking on KM initiatives.
ducted with Most of the 150
key respondents were males (57.3%)
personnel in
tives in these and in the age group of 21 to 29 years (50.7%). A
organizations. A
executives from 17
majority of the respondents public
had university degrees
interviewed. (86.7%), with the rest having high school educa
tion. Their work covered functional areas such as
Of the 17 public organizations contacted, 10 information systems (33.3%), corporate services
organizations (covering 7 industries) were willing (10%), marketing (8.7%), strategic planning
to participate in the survey. The organizations (8.7%), human resource (7.3%), and research and
were offered a report of our findings as an development (7.3%). On average, the respon
incentive to participate. A senior KM executive dents had work experience of 7.5 years and had

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Table 4. Characteristics of Participating Organizations

Industry Number of Responses Percentage of Responses
Defense 23 15.3%
Education 11 7.3%
Information Technology 61 40.7%
Library 6 4.0%
Real Estate and Construction 17 11.3%
Rehabilitation 8 5.3%
Research and Development 24 16.0%
Total_150_100.0%_
the organizational
been using EKRs as knowledge contributors for mechanisms sur
KM initiatives.
2.5 years. They used information technology These similaritie
extensively in the course of their responses
work. from the 10 organization
for data analysis.
All organizations surveyed used EKRs as a critical
component of their KM initiative. For instance one
organization in the education sector used a Lotus
Data
Notes-based EKR to store student Analysis
counseling and Resu
case studies and project reviews. Another organi
zation in the real estate sector The
usedconstructs
their Lotus were first assessed
Notes-based system for storing andcase studies,
validity. After ascertaining that
project reviews, and lessons learned
could (after
meetaction
parametric requirement
reviews). All EKRs studied were sion
mainly test, theto
used hypotheses were
store project reviews, case studies,
moderated lessons
multiple regression anal
learned, and best practices. The contents
tical testsof the carried out at a 5 p
were
significance.
EKRs were in the form of documents (Word or pdf
format) and presentations (PowerPoint format).
Keywords and other metadata were used to index
and retrieve the content. Therefore, the contents
Reliability
were fairly structured. The number of EKR and Validity
users
in these organizations varied from about 120 to
The constructs
2,000, while the organization sizes ranged were assessed
from for reliability using
Cronbach's
about 250 to 2,500. All organizations alpha (Cronbach
provided KM 1951). Nunnally
(1978) suggested
training and incentives for knowledge that a value of at least 0.70
sharing.
indicated adequate
There was also top management support for reliability.
KM In order to improve
the reliabilities
initiatives. However, there were no explicit man of the corresponding constructs,
one question
dates for employees to use EKRs was omitted
(or other KMfrom each of the
following constructs:
systems), i.e., EKR usage was voluntary. organizational reward
(OREW1), reciprocity
Information obtained and observations of EKRs (RECP1), pro-sharing norms
during interviews revealed that (PSNM5),
all and
of identification
the 10 (IDEN6). Subse
quently, all of
organizations participating in this survey werethe constructs had adequate
reliability
comparable in terms of their EKR (see Table 5).and
capabilities

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Table 5. Reliability of Constructs

Construct Number of Questions Cronbach's Alpha
Loss of knowledge power (LOKP) 4 0.95
Codification effort (CEFF) 5 0.85
Organizational reward (OREW) 4 0.96
Image (IMAG) 4 0.89

Reciprocity (RECP) 3 0.85
Knowledge self-efficacy (KSEF) 4 0.96
Enjoyment in helping others (EHLP) 4 0.96

Generalized trust (GTRU) 4 0.85
Pro-sharing norms (PSNM) 4 0.93
Identification (IDEN) 6 0.96
EKR usage (EUSG) 3 0.85
The questions were tested for validity using factor Hypotheses Tests
analysis with principal components analysis and
varimax rotation. Convergent validity was as Studies in information systems (e.g., McKeen etal.
1994; Weill and Olson 1989) and in other disci
sessed by checking loadings to see if items within
the same construct correlate highly amongst plines (e.g., Jehn et al. 1999) have used moder
themselves. Discriminant validity was assessed ated multiple regression to test interaction effects.
by examining the factor loadings to see if ques Moderated multiple regression is a hierarchical
tions loaded more highly on their intended procedure that first tests the relationship between
constructs than on other constructs (Cook and independent constructs and the dependent con
Campbell 1979). Loadings of 0.45 to 0.54 are struct, and then tests the relationship between
considered fair, 0.55 to 0.62 are considered good, interaction terms and the dependent construct
0.63 to 0.70 are considered very good, and above (Sharma et al. 1981; Stone and Hollenbeck 1984).
0.71 are considered excellent (Comrey 1973). Interaction terms are computed by multiplying two
independent constructs. A significant change in
Factor analysis yielded 11 components with explanatory power between the two steps, which
eigenvalues above 1 (see Table 6). These 11 can be assessed by looking at the significance of
components corresponded to the 11 constructs. the change in F value, indicates the presence of
Two questions for codification effort (CEFF4 and moderating effects.
CEFF5) tapped onto other constructs and were
omitted. All other questions had at least good In this study, the independent constructs were
loadings on their intended constructs. After entered in the first step of regression and the
omitting the two questions, the reliability of the interaction terms were added in the second step.
codification effort construct improved to 0.91. All interaction terms were assessed simulta

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
I Table 6. Validity of Questions

_^___^_Factor_
Question 1|2|3|4|5|6|7|8|9|10|11
LOKP1 -0.08 -0.20 0.03 -0.15 0.87 0.01 -0.06 0.00 -0.05 -0.02 -0.08
LOKP2 -0.16 -0.23 0.00 -0.13 0.90 -0.04 0.04 -0.04 -0.04 0.00 0.00
LOKP3 -0.13 -0.17 0.01 -0.08 0.90 0.00 0.00 -0.09 -0.11 -0.06 -0.04
LOKP4_-0.16 -0.23 0.09 -0.10 Q.87 0-03 0.03 -0.11 -0.09 0.00 -0
CEFF1 -0.13 -0.15 0.06 0.01 0.09 -0.05 -0.06 0.85 -0.15 -0.19 -0.09
CEFF2 -0.19 -0.14 -0.08 -0.08 0.10 0.00 -0.17 0.89 -0.10 -0.04 -0.04
CEFF3 -0.14 0.00 -0.07 -0.11 0.04 0.04 -0.15 0.90 -0.12 -0.02 -0.08
CEFF4 -0.08 -0.10 -0.13 -0.09 0.14 -0.07 -0.12 0.40 -0.11 -0.15 -0.42
CEFF5_-0.06 -0.09 -0.15 -0.12 0.16 -0.09 -0.11 Q.39 -0.10 -0.13 -0.43
OREW2 0.03 0.03 0.88 0.11 0.01 0.28 0.04 0.01 0.09 0.10 0.14
OREW3 0.07 0.00 0.91 0.06 0.06 0.24 0.06 -0.02 0.12 0.11 0.09
OREW4 0.07 0.05 0.92 0.08 0.08 0.20 0.04 -0.01 0.07 0.11 0.11
OREW5_0.07 0.03 Q.89 -0-05 -0.01 0.13 0.01 0.10 0.04 0.1
IMAG1 0.13 0.21 0.37 -0.08 -0.05 0.69 0.09 0.03 -0.11 0.18 0.07
IMAG2 0.03 0.03 0.19 0.00 0.18 0.82 0.03 -0.01 0.25 0.10 0.05
IMAG3 0.12 0.22 0.26 0.04 -0.07 0.78 0.13 -0.02 -0.09 0.18 0.04
IMAG4_0.07 0.15 0.24 0.01 -0.08 Q.86 0-04 0.02 -0.04 0.13 0
RECP2 0.19 0.16 0.13 0.10 -0.08 0.10 -0.03 0.15 0.33 0.71 0.15
RECP3 -0.01 0.09 0.21 0.03 0.02 0.21 0.02 0.03 0.01 0.87 0.02
RECP4_0.10 0.01 0.13 0.07 0.01 0.18 -0.02 0.11 0.07 Q
KSEF1 0.16 0.21 0.01 0.87 -0.11 -0.04 0.10 0.04 0.02 0.00 0.09
KSEF2 0.10 0.19 0.01 0.89 -0.11 0.03 0.11 0.07 0.08 0.07 0.15
KSEF3 0.06 0.09 0.07 0.92 -0.13 0.04 0.08 0.04 0.05 0.09 0.12
KSEF4_0.03 0.11 0.05 Q.91 -0-11 -0-03 0.06 0.04 0.03 0
EHLP1 0.15 0.82 -0.02 0.16 -0.24 0.16 0.03 0.18 0.08 0.11 0.15
EHLP2 0.18 0.86 -0.02 0.18 -0.27 0.14 0.04 0.11 0.12 0.08 0.12
EHLP3 0.22 0.84 0.08 0.14 -0.26 0.15 0.04 0.04 0.12 0.03 0.08
EHLP4_0-16 Q.81 0.08 0.24 -0.27 0.15 0.08 0.02 0.07 0.0
GTRU1 0.37 0.00 0.20 0.05 -0.11 -0.01 0.25 0.18 0.58 0.09 0.11
GTRU2 0.23 0.16 0.09 0.00 -0.15 -0.02 0.18 0.12 0.74 0.17 -0.06
GTRU3 0.29 0.19 0.07 0.10 -0.06 0.08 0.19 0.09 0.79 0.08 0.04
GTRU4_0.31 0.01 0.05 0.09 -0.08 -0.05 0.37 0.17 Q.63 0-
PSNM1 0.36 0.04 0.03 0.14 -0.08 0.11 0.81 0.08 0.24 -0.07 0.01
PSNM2 0.35 0.07 0.05 0.18 -0.06 0.10 0.82 0.10 0.19 -0.05 0.03
PSNM3 0.42 0.09 0.03 0.08 0.01 0.06 0.70 0.20 0.25 0.05 0.09
PSNM4_0.43 0.01 0.09 0.07 0.04 0.09 Q.66 0-24 0.25
IDEN1 0.80 0.16 0.10 0.02 -0.15 0.06 0.22 0.11 0.07 0.09 0.12
IDEN2 0.86 0.16 0.13 0.05 -0.13 0.03 0.16 0.11 0.13 0.07 0.08
IDEN3 0.84 0.15 0.11 0.05 -0.08 0.09 0.19 0.07 0.15 0.04 -0.03
IDEN4 0.87 0.16 0.03 0.07 -0.14 0.05 0.19 0.13 0.16 0.06 -0.04
IDEN5 0.82 0.04 -0.01 0.06 -0.09 0.07 0.18 0.13 0.27 0.02 0.02
IDEN7 0.85 0.10 -0.06 0.17 -0.08 0.06 0.15 0.01 0.15 0.04 0.10
EUSG1 Ô04~ -0.03 ~O06" 0.23 -0.04 ~oTj 013" 0.10 0.05 -0.05 0.
EUSG2 0.18 0.38 0.25 0.19 -0.06 0.07 0.04 0.09 0.03 0.23 0.63
EUSG3_0.21 0.42 0.21 0.20 -0.10 0.05 0.00 0.08 -0.0
Eigenvalue_5.70 3.98 3.85 3.76 3.67 2.97 2.87 2.69 2.
Variance 13.16 9.27 8.96 8.75 8.54 6.92 6.68 6.27 6.19 5.59 3.88
explained (%)_
Cumulative 13.16 22.43 31.39 40.14 48.68 55.60 62.28 68.55 74.74 80.33 84.21
variance (%)

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Table 7. Results of Hypothesis Tests

Standardized
Coefficient Hypothesis Test
Step 1: Main Effects
Loss of knowledge power (LOKP) 0.05
Codification effort (CEFF) -0.07
Organizational reward (OREW) 0.22**
Image (IMAG) -0.05
Reciprocity (RECP) 0.11
Knowledge self-efficacy (KSEF) 0.25*** H6 was supported
Enjoyment in helping others (EHLP) 0.43*** H7 was supported
Generalized trust (GTRU) -0.13
Pro-sharing norms (PSNM) 0.04
Identification (IDEN) 0.04
R2 0.44
Adjusted R2 0.38
F 10.95***
Step 2: Interaction Terms

LOKP^PSNM -0.11 H1 was not supported
CEFF-GTRU -0.18* H2a was supported
CEFF-PSNM 0.01 H2b was not supported
CEFFHDEN 0.07 H2c was not supported
OREW-PSNM -0.17 H3a was not supported
OREWHDEN 0.23** H3b was not supported
IMAG'PSNM 0.12 H4 was not supported
RECP?PSNM -0.18* H5 was supported
R2 0.52
Adjusted R2 0.45
F 7.72***
R2 change 0.08
F change 3.23**
*p < 0.05; **p < 0.01, ***p < 0.001
neously so that their effects

butors.7could be seen in
The change in
context of the overall steps
model (i.e.,
of in the prese
regression was
p < 0.01),
of other main and interaction indicating
effects). t
To allev
possible collinearity problems, the (i.e.,
second step values of a
testin
constructs were centered (mean
could be subtracte
interpreted (se
during regression (Aiken and West 1991). The
value of 0.52 and adjusted R2 value of 0.45
7.72, p < 0.001) indicated that the overall mod
was more than satisfactory in explaining
7Falk and Miller (1992) indicate that explanatory power
variance in EKR usage by
(R2 value) greater than knowledge
10 percent is acceptable. con

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Table 7 summarizes the results of hypotheses community size) were included in a moderated
tests. Loss of knowledge power had no significant multiple regression model together with the 11
relationship with EKR usage even under conditions original constructs. The results demonstrated that
of weak pro-sharing norms, i.e., H1 was not sup the significant main effects and interaction terms
ported. Codification effort had a significant nega remained the same as in Table 7. None of the
tive relationship with EKR usage under conditions control variables had a significant impact on EKR
of weak generalized trust but not under conditions usage. Also, the inclusion of the control variables
of weak pro-sharing norms and weak identification, did not significantly increase the variance ex
i.e., H2a was supported but H2b and H2c were not plained. Therefore, the results of hypotheses tests
supported. Organizational reward had a significant (see Table 7) appeared to be stable and
positive relationship with EKR usage, not under independent of control variables.
conditions of weak pro-sharing norms but under
conditions of strong identification (this was oppo
site to the prediction of hypothesis H3b), i.e., H3a
and H3b were not supported. Image had no Discussion and Implications WM
significant relationship with EKR usage even under
conditions of weak pro-sharing norms. Hence, H4 Based on our findings, loss of knowledge power
was not supported. Reciprocity had a significant did not significantly affect EKR usage by knowl
positive relationship with EKR usage under con edge contributors, not even under conditions of
ditions of weak pro-sharing norms, i.e., H5 was weak pro-sharing norms. This may be due to the
supported. Knowledge self-efficacy and enjoyment fact that knowledge contribution is voluntary in the
in helping others had significant positive relation organizations surveyed in this study. Under such
ships with EKR usage, i.e., H6 and H7 were circumstances, knowledge contributors can decide
supported. what to contribute to EKRs. Hence, they need not
fear that their knowledge contribution would render
The standardized coefficients in Table 7 indicate them less valuable to the organization. The
that, relatively, enjoyment in helping others had the respondents of this survey are actively partici
strongest impact on EKR usage by knowledge pating in the creation or acquisition of new knowl
contributors followed by knowledge self-efficacy edge in the course of their work. Thus, they can
and organizational rewards (both directly and remain valuable to the organization even after
moderated by identification). Codification effort contributing knowledge to EKRs. Another plau
moderated by generalized trust and reciprocity sible explanation for this result is self-selection
moderated by pro-sharing norms had the least among the survey respondents. Knowledge contri
impact among the significant determinants of EKR butors who responded to this survey may also be
usage by knowledge contributors. those who are more likely to contribute their
knowledge to EKRs. These knowledge contri
butors may tend to be less concerned about the
loss of knowledge power.
Control Variables
The relationship between codification effort and
Further analysis was carried out to make sure the EKR usage by knowledge contributors was
significant results were not due to covariation with contingent on generalized trust. As hypothesized,
control variables. Previous literature suggests that when generalized trust is strong, codification effort
gender (Jarvenpaa and Staples 2000), age may not be a deterrent for EKR usage by knowl
(Jarvenpaa and Staples 2000), work experience edge contributors. However, the relationship
(Constant et al. 1994), and education (Constant et between codification effort and EKR usage by
al. 1994) may affect knowledge contribution knowledge contributors was not contingent on pro
behavior. EKR user community size may also sharing norms and identification. Therefore, even
influence EKR usage. These control variables when there are norms of collaboration and
(age, gender, education, work experience, and cooperation or when the interests of people are

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
aligned with those of the organization, such when pro-sharing norms are weak. Prior literature
contextual factors do not impact the relationship suggests that increased recognition by colleagues
between codification effort and EKR usage by or the organizational community can be an impor
knowledge contributors. The existing literature (Ba tant motivator for employees to contribute their
et al. 2001; Goodman and Darr 1998) has dis knowledge (Constant et al. 1994; Hall 2001;
cussed the deterrent effect of codification effort on Kollock 1999; O'Dell and Grayson 1998). How
knowledge contribution. The findings of this study ever, ourfindings suggest that future research may
extend the previous literature by revealing that the want to further examine the role of pro-sharing
relationship between codification effort and norms in relation to image as a motivator for
knowledge contribution to EKRs is salient when knowledge contribution.
generalized trust is weak.
The relationship between reciprocity and EKR
The relationship between organizational reward usage by knowledge contributors was contingent
and EKR usage by knowledge contributors was on pro-sharing norms. As hypothesized, when
both direct and contingent on identification. pro-sharing norms are strong and there is a
Contrary to hypothesis H3b, this relationship climate of collaboration and cooperation, knowl
appears to be stronger when identification is edge contributors do not look for reciprocity when
strong, i.e., when knowledge contributors to EKR contributing their knowledge to EKRs. However,
share the same interests as the organization, they when pro-sharing norms are weak, reciprocity
tend to be more motivated by organizational benefit is a motivator for knowledge contribution to
rewards. It appears that if knowledge contributors EKRs. This finding extends prior literature
do not share the interests of the organization, even (Connolly and Thorn 1990; Kollock 1999) by
organizational reward may not motivate them to indicating that the relationship between reciprocity
contribute their knowledge to EKRs. The relation and EKR usage by knowledge contributors is
ship between organizational reward and EKR significant when pro-sharing norms are weak.
usage was not contingent on pro-sharing norms.
Several organizations have used organizational Knowledge self-efficacy significantly impacted EKR
reward (Ba et al. 2001; Beer and Nohria 2000; Hall usage by knowledge contributors. As hypothe
2001) to build up pro-sharing norms among their sized, when people are confident of their ability to
employees. Being used to obtaining organiza contribute knowledge that would be useful to the
tional reward, knowledge contributors may organization, they tend to be more motivated to do
continue to expect such rewards for knowledge so through EKRs. This result is consistent with
contribution to EKRs even after pro-sharing norms previous KM experiments (Constant et al. 1996)
have developed. The findings of this study extend and conceptual articles (Ba et al. 2001). Enjoy
prior literature by revealing that the relationship ment in helping others also significantly affected
between organizational reward and EKR usage by EKR usage by knowledge contributors. As hypoth
knowledge contributors is most significant when esized, when people feel good about contributing
identification is strong. knowledge to help others, they tend to be more
motivated to do so through EKRs. Again, this
Image did not significantly affect EKR usage by result is consistent with previous KM conceptual
knowledge contributors, not even under conditions (Ba et al. 2001) and case study literature
of weak pro-sharing norms. This may be due to (Davenport and Prusak 1998) highlighting altruism
the dual effects of pro-sharing norms. On one as a motivator for knowledge sharing.
hand, strong teamwork and collaboration norms
may reduce the need for improved image as a
motivator for knowledge contribution to EKRs. On
the other hand, strong error tolerance and diversity Implications for Theory
norms may enhance the need for this benefit since
the risk of making mistakes during contribution are This study advances theoretical development in
lessened. The converse dual effects may occur the area of KM in general and EKRs in particular.

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
??^
Costs
Codification effort
Generalized Trust \^^

i-:-1 -?-18*
Extrinsic Benefits \^
Organizational Reward- ^^^l-1
n 22**
-^ EKR USAGE
Organizational Reward _0 23**
Identification ^ BY
I-:-1>Tma*-~ ~* KNOWLEDGE
Reciprocity __^? "u-'?
Pro-sharing Norms ^^ CONTRIBUTORS
- 0.25*** s'
Intrinsic Benefits
I-1 ^ /^ R2 = 0.52
I-1 ^ 0.43*** ^^ ^
Knowledge Self-efficacy ^ ^
Enjoyment in Helping '

Others
^????^?_^_^^____^______________________________________________________
Figure 2. The Revised Research Model
It demonstrates that cost Theand benefit

impacts factors
of extrinsic be
rewards and reciprocity)
derived from social exchange theory moderated by ap
contextual factors derived
by from social factors
contextual capital (id
theory can predict EKR usage
sharing
by norms,
knowledgerespectively
con
tributors. Based on the findings of this
the provision study,
of we ben
extrinsic
refine our original research model
adequate (see Figure
motivators of1)know
and propose an alternative research
EKRs modelextrinsic
unless these that
in usage
can better account for EKR appropriate contexts. F
by knowledge
contributors (see Figure tional reward
2). The seems to wo
explanatory
power of the revised research model
fication demonstrates
is strong while rec
the value of using social exchange
most theory
effective whenandpro-sh
However
social capital theory to predict the
usage effects
of of int
collective
technologies such as EKRs. edge self-efficacy and enjoym
appear to be direct. Given t
are light
The results of this study shed sought
on as
howends desired b
extrinsic
or intrinsic benefits may factors
differ in
do terms of atheir
not play signific
the value of
impact on EKR usage by knowledge these benefits
contributors.

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
butors. The fostering of intrinsic benefits alone edge self-efficacy among valued knowledge
may be sufficient to motivate knowledge contri contributors by indicating to them that their
butors to contribute their knowledge to EKRs, in knowledge contribution makes a significant dif
many contexts (different combinations of con ference to the organization. This can be done by
textual factors). The impact of cost factors (codifi highlighting the improved organizational perfor
cation effort) appears to be moderated by mance arising from their knowledge contributions.
contextual factors (generalized trust). This implies Organizations such as Amazon.com regularly
that measures to alleviate the costs of knowledge recognize their top reviewers, serving as a way to
contribution to EKRs may only be necessary in enhance the self-efficacy of these knowledge
specific contexts. For example, it may be useful to contributors.
reduce codification effort when generalized trust is
weak but it may not be necessary to do this when Second, management can attempt to raise the
generalized trust is strong. In summary, three level of enjoyment that knowledge contributors
contextual factors that have been found to be experience as they help others. This may be done
important are generalized trust, pro-sharing norms, by connecting knowledge contributors and knowl
and identification. edge recipients in order to allow recipients to
express their appreciation for the knowledge
Apart from identifying factors that determine EKR received. The realization that their colleagues
usage by knowledge contributors, this study also have benefitted from their knowledge contribution
contributes to theory by unveiling factors that do can increase the feeling of altruism among
not appear to impact EKR usage by knowledge knowledge contributors (Davenport and Prusak
contributors. For example, the loss of knowledge 1998). As a way of motivating knowledge contri
power is thought to be a barrier to knowledge butors, knowledge seekers can be rewarded for
contribution. However, in our study, this is not a finding solutions from EKRs and acknowledging
significant concern for knowledge contributors to the sources of the solutions. The Most Valuable
EKR. As another example, image is considered as Professionals Program at Microsoft Corporation is
a motivator for knowledge contribution. However, an example of an initiative that raises altruism (and
our findings show that knowledge contributors to community spirit) in this way. Through this pro
EKR may not be concerned about the image gram, people who have provided useful technical
associated with knowledge contribution. These assistance to other users of Microsoft technology
results suggest that future research should take a are identified and informed that they have helped
closer look at how power and image are perceived others (Microsoft 2002).
by knowledge contributors. Besides identifying the
constructs that can or cannot predict EKR usage Third, organizational reward (such as better work
by knowledge contributors, this study also under assignment, promotion incentive, salary incentive,
takes a rigorous conceptual and empirical process bonus incentive, or job security) seems to be
to develop measures for each of these constructs. effective for encouraging EKR usage by knowl
edge contributors. Organizations have used
various forms of organizational reward to
encourage employees to contribute their knowl
Implications for Practice edge to EKRs. For example, IBM Global Services
introduced schemes to identify and reward specific
Collectively, the results of this study indicate the instances of knowledge contribution (Berry 2000).
circumstances under which organizational mea Organizational reward appears to be particularly
sures to promote knowledge contribution to EKRs effective under conditions of strong identification.
may be more effective. These results offer sug Thus, to build a critical mass of knowledge
gestions to management about how to promote contributors, management can offer organizational
EKR usage by knowledge contributors. First, reward and publicize such reward first among
management can raise the perceptions of knowl employees, groups, or business units whose

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
interests are known to align with those of the other's knowledge. Such practices are adhered to
organization. in exemplar KM organizations such as Buckman
Laboratories (Buckman 2004).
Fourth, management can raise the perceptions of
reciprocity benefit among knowledge workers by The fact that enjoyment in helping others is the
highlighting situations where requests for help from most important motivator for contributors to EKRs
knowledge contributors have been promptly followed by knowledge self-efficacy and organi
answered. Valued knowledge contributors can be zational rewards (both directly and moderated by
asked to testify in KM events about how they have identification) implies a priority for the managerial
benefitted from the knowledge contribution of recommendations outlined above. For example,
others. Reciprocity appears to be particularly measures to increase enjoyment in helping others
important when pro-sharing norms are weak. may be more effective in terms of encouraging
Alternatively, management can strengthen pro EKR contributors than organizational rewards and,
sharing norms to reduce the necessity of reci therefore, should be given higher priority.
procity benefit for knowledge contributors to EKR. Similarly, the measures to increase reciprocity
Organizations have successfully promoted pro benefits and reduce codification effort may be of
sharing norms through a variety of means. For lower priority in motivating contributions to EKRs.
example, at British Petroleum open office spaces
helped employees to more easily consult each
other (Chiem 2001). General Electric transfers its
employees between departments to promote pro Limitations and Future Studies
sharing norms (Dzinkowski 2001).
Results of this study must be interpreted in the
Finally, management can reduce codification effort context of its limitations. First, the use of cross
by deploying KM systems (including EKRs) that sectional data and regression analysis do not allow
facilitate entry of knowledge and thereby reduce the possibility of bidirectional (feedback) effects to
the time and effort needed to codify knowledge. be explored. For instance, the effects of EKR
Commercially available KM systems provide usage on subsequent perceptions of cost and
capabilities that prompt for knowledge and auto benefit factors by knowledge contributors has been
matically organize the knowledge captured to recognized but cannot be examined. Future
reduce codification effort. Examples are taxonomy studies can collect longitudinal data to assess
generators, classifiers, and clustering engines such bidirectional (feedback) effects.
available from companies such as Autonomy and
Verity. KM systems may be designed to allow Second, based on a sample of 150 respondents,
more natural forms of knowledge contribution (e.g., several significant results have been obtained.
audio or video) as opposed to purely text However, a larger sample that brings more statis
contributions. Besides trying to reduce the time tical power would have allowed more sophisticated
and effort needed to codify knowledge, manage statistical analysis. With such samples, future
ment can also allocate time for employees to share studies can test a second-order model using struc
knowledge by integrating this activity into regular tural equation modeling techniques. Cost and
work processes. Such a practice has been benefit can be modeled as second-order con
common in consultancy firms like Accenture structs with the various cost and benefit factors as
(Hansen et al. 1999). Codification effort appears formative indicators. Such a model would allow a
to be a salient barrier when generalized trust is more rigorous test of the constructs from social
weak. Hence, when it is difficult to deploy KM exchange theory.
systems that facilitate codification efforts, manage
ment can raise the level of generalized trust. This Third, our research model was empirically tested
can be done by giving due credit for knowledge based on the responses of knowledge contributors
contributions and ensuring appropriate usage of from 10 public organizations in Singapore. Since

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
the manner of operation and culture of public and tual) factors, this model
private (for-profit) organizations in Singapore is not power when compared to
substantially different, the results of our study are broader sense, this study d
potentially generalizable to private organizations of using social exchange th
with KM initiatives. However caution must be theory to account for th
exercised when attempting to generalize the technologies (which include
results across a range of organizations operating KM systems). Besides con
in varied contexts. Future studies can replicate building in the area of KM
this study using our revised research model (see particular, the results of th
Figure 2) in other contexts. For example, a similar implications to KM practitio
research approach can be used to investigate EKR
usage by knowledge seekers. Other forms of KM In a future characterized by
systems (such as those supporting the personali effective leverage of org
zation strategy) can also be studied in a similar would be a factor different
way. In addition, our revised research model can from less successful org
be tested with respondents from different organi toward facilitating knowled
zational settings to assess the external validity of contributions to EKRs need
the results. organizations invest m
initiatives, it is imperative
Fourth, future research can extend our revised initiatives and KM systems
theoretical model (see Figure 2) to account for the continue to generate finding
remaining unexplained variance in EKR usage by
knowledge contributors. To extend the revised
theoretical model, additional theoretical perspec
References
tives such as the technology acceptance model
(Davis 1989) and the task-technology fit model
Adler, P. S. "Market, Hier
(Goodhue and Thompson 1995) may be incor
Knowledge Economy a
porated. The technology acceptance model can
Capitalism," Organization
help to better account for the ease of use and
pp. 215-234.
usefulness of EKRs. The task-technology fit
Agarwal, R. "Individual Acce
model can help to better explain the effects of
Technologies," in Framin
knowledge type on knowledge contribution to
Management: Projecting th
EKRs. Additionally, the use of concepts from the Past, R. W. Zmud (Ed.),
organizational memory information systems litera Resources, Cincinnati, O
ture (e.g., Stein and Zwass 1995) may enable Aiken, L. S., and West, S. G
researchers to investigate specific subsystems of Testing and Interpretin
EKRs as well as the KM activities associated with
Publications, Newbury Par
these subsystems. Alavi, M. "Managing Organ
in Framing the Domains
Projecting the Future thr
Zmud (Ed.), Pinnaflex Ed
Conclusion i Cincinnati, OH, 2000, pp.
Alavi, M., and Leidner, D. E
This study develops and tests
menta theoretical
Systems: model
Issues, C
that explains how cost and benefit
fits," factors can of
Communications
37.
impact EKR usage by knowledge contributors, and
the contexts under which these
Alavi, M., and Leidner, D. effects may
E. "Review: Knowledge
operate. By encompassing Management
individual and Knowledge Management
level (cost
Systems: Conceptual level
and benefit) factors and community Foundations and Re
(contex

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
search Issues," MIS Quartedy (25:1), 2001, pp. Cohen, D., and Prusak, L. In Good Company:
107-136. How Social Capital Makes Organizations Work,
Ba, S., Stallaert, J., and Whinston, A. B. Harvard Business School Press., Boston, 2001
"Research Commentary: Introducing a Third Coleman, J. Foundations of Social Theory,
Dimension in Information Systems Design?The Harvard University Press, Cambridge, MA,
Case for Incentive Alignment," Information Sys 1990.
tems Research (12:3), 2001, pp. 225-239. Comrey, A. L. A First Course in Factor Analysis,
Babcock, P. "Shedding Light on Knowledge Academic Press, NewYork, 1973.
Management," HR Magazine (49:5), 2004, pp. Connolly, T., and Thorn, B. K. "Discretionary
46-50. Databases: Theory, Data, and Implications," in
Bandura, A. Social Foundations of Thought and Organizations and Communication Technology.
Action: A Social Cognitive Theory, Prentice J. Fulk and C. Steinfield (Eds.), Sage Publi
Hall, Englewood Cliffs, NJ, 1986. cations, Newbury Park, CA, 1990, pp. 219-233.
Baumeister, R. F. "A Self-Presentational View of Constant, D., Kiesler, S., and Sproull, L. "What's
Social Phenomena," Psychological Bulletin Mine Is Ours or Is It? A Study of Attitudes
(91:1), 1982, pp. 3-26. about Information Sharing," Information
Beer, M., and Nohria, N. "Cracking the Code of Systems Research (5:4), 1994, pp. 400-421.
Change," Harvard Business Review (78:3), Constant, D., Sproull, L., and Kiesler, S. "The
2000, pp. 133-141. Kindness of Strangers: The Usefulness of
Electronic Weak Ties for Technical Advice,"
Berry, J. "Employees Cash in on KM?Knowledge
Organization Science (7:2), 1996, pp. 119-135.
Management Programs Pay Rewards to Share
Cook, M., and Campbell, D. T. Quasi-Experimen
Ideas," Internetweek (814), 2000, pp. 45-46.
tation: Design and Analysis Issues for Field
Blau, P. M. Exchange and Power in Social Life,
Settings, Houghton Mifflin, Boston, 1979.
John Wiley, NewYork, 1964.
Cronbach, L. J. "Coefficient Alpha and the Internal
Bock, G. W., and Kim, Y. G. "Breaking the Myths
Structure of Tests," Psychometrika (16), 1951,
of Rewards: An Exploratory Study of Attitudes
pp. 297-334.
about Knowledge Sharing," Information
Cross, R., and Baird, L. "Technology Is Not
Resource Management Journal (15:2), 2002,
Enough: Improving Performance by Building
pp. 14-21.
Organizational Memory," Sloan Management
Bock, G. W., Zmud, R. W., Kim, Y. G., and Lee,
Review (4V.3), 2000, pp. 69-78.
J. N. "Behavioral Intention Formation in Knowl
Davenport, T. H., De Long, D. W., and Beers, M.
edge Sharing: Examining the Roles of Extrinsic
C. "Successful Knowledge Management Pro
Motivators, Social-Psychological Forces, and jects," Sloan Management Review (39:2), 1998,
Organizational Climate," MIS Quarterly (29:1), pp. 43-57.
March 2005, pp. 87-111. Davenport, T. H., and Prusak, L. Working Knowl
Brown, J. S., and Duguid, P. "Organizational edge: How Organizations Manage What They
Learning and Communities of Practice," Organi Know, Harvard Business School Press, Boston,
zation Science (2:1), 1991, pp. 40-57. 1998.
Buckman, R. H. Building a Knowledge-Driven Davis, F. D. "Perceived Usefulness, Perceived
Organization, McGraw Hill, NewYork, 2004. Ease of Use, and User Acceptance of Informa
Cheney, G. "On the Various and Changing tion Technology," MIS Quarterly (13:3), 1989,
Meanings of Organizational Membership: A pp. 319-341.
Field Study of Organizational Identification," Deci, E. L., and Ryan, R. M. "The Empirical
Communication Monographs (50), 1983, pp. Exploration of Intrinsic Motivational Processes."
342-362. in Advances in Experimental Social Psychology.
Chiem, P. X. "Form Follows Strategy," Knowledge L. Berkowitz (Ed.), Academic Press, NewYork,
Management Magazine, October 2001 (avail 1980, pp. 39-80.
able online at http://www.destinationkm.com/ Dooley, D. Social Research Methods, Prentice
articles/default.asp?ArticlelD=369). Hall, Upper Saddle River, NJ, 2001.

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Dzinkowski, R. "Removing Boundaries to Hickson, D. J., Hinings, C. R., Lee, C. A., Schenk,
Learning," Knowledge Management Magazine, R. E., and Pennings, J. M. "A Strategic Con
May 2001 (available online at http://www. tingencies Theory of Interorganizational Power,"
destinationkm.com/articles/default.asp7Article Administrative Sciences Quarterly (16:2), 1971,
ID=240). pp. 216-229.
Falk, R. F., and Miller, N. B. A Primer for Soft Igbaria, M., Parasraman, S., and Baroudi, J. J. "A
Modeling, University of Akron Press, Akron, Motivational Model of Microcomputer Usage,"
OH, 1992. Journal of Management Information Systems
Fulk, J., Flanagin, A., Kalman, M., Monge, P. R., (13:1), 1996, pp. 127-143.
and Ryan, T. "Connective and Communal Jarvenpaa, S. L., and Staples, D. S. "The Use of
Public Goods in Interactive Communication Collaborative Electronic Media for Information
Systems," Communication Theory (6:1), 1996, Sharing: An Exploratory Study of Determi
pp. 60-87. nants," Journal of Strategic Information Sys
Goodman, P. S., and Darr, E. D. "Computer tems (9:2/3), 2000, pp. 129-154.
Aided Systems and Communities: Mechanisms Jehn, K. A., Northcraft, G. B., and Neale, M. A.
for Organizational Learning in Distributed (1999). "Why Differences Make a Difference: A
Environments," MIS Quarterly (22:4), 1998, pp. Field Study of Diversity, Conflict, and Perfor
417-440. mance in Workgroups," Administrative Science
Goodhue, D. L., and Thompson, R. L. "Task Quarterly (44:4), 1999, pp. 741-763.
Technology Fit and Individual Performance," Johnson, W. L., Johnson, A. M., and Heimberg, F.
MIS Quarterly (19:2), 1995, pp. 213-236. "A Primary and Second Order Component
Gray, P. H. "The Impact of Knowledge Reposi Analysis of the Organizational Identification
tories on Power and Control in the Workplace," Questionnaire," Educational and Psychological
Information Technology and People (14:4), Measurement (59:1), 1999, pp. 159-170.
2001, pp. 368-384. Kalman, M. E. The Effects of Organizational
Green, G. I. "Perceived Importance of Systems Commitment and Expected Outcomes on the
Analyst's Job Skills, Roles and Non-Salary Motivation to Share Discretionary Information in
Incentives," MIS Quarterly (13:2), 1989, pp. a Collaborative Database: Communication
115-133. Dilemmas and Other Serious Games, unpub
Grover, V., and Davenport, T. H. "General Per lished Ph.D. Thesis, University of Southern
spectives on Knowledge Management: Fos California, Los Angeles, CA, 1999.
tering a Research Agenda," Journal of Manage Kim, J., and Mueller, C. W. Factor Analysis:
ment Information Systems (18:1), 2001, pp. 5 Statistical Methods and Practical Issues, Sage
21. Publications, Newbury Park, CA, 1981.
Hall, H. "Social Exchange for Knowledge Kollock, P. "The Economies of Online Coopera
Exchange," paper presented at Managing tion: Gifts and Public Goods in Cyberspace," in
Knowledge: Conversations and Critiques, Uni Communities in Cyberspace, M. Smith and P.
versity of Leicester Management Centre, April Kollock (Eds.), Routledge, NewYork, 1999, pp.
10-11,2001 (available online at http://www.soc. 220-239.
napier.ac.uk/publication/op/getpublication/ Krebs, D. L. "Empathy and Altruism," Journal of
publicationid/321908). Personality and Social Psychology (32:6), 1975,
Hansen, M. T., Nohria, N., and Tierney, T. pp. 1132-1146.
"What's Your Strategy for Managing Knowl Leonard-Barton, D. Wellsprings of Knowledge:
edge?," Harvard Business Review (77:2), 1999, Building and Sustaining the Source of Innova
pp. 106-116. tion, Harvard Business School Press, Boston,
Hargadon, A. B. "Firms as Knowledge Brokers: 1995.
Lessons in Pursuing Continuous Innovation," Liebowitz, J., and Beckman, T. Knowledge
California Management Review (40:3), 1998, Organizations: What Every Manager Should
pp. 209-227. Know, CRC Press, Boca Raton, FL, 1998.

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Markus, M. L. "Towards a Theory of Knowledge Internal Best Practices," California Management

Reuse: Types of Knowledge Reuse Situations Review (40:3), 1998, pp. 154-174.
and Factors in Reuse Success," Journal of O'Reilly, C, and Chatman, J. "Organizational
Management Information Systems (18:1), 2001, Commitment and Psychological Attachment:
pp. 57-94. The Effects of Compliance, Identification, and
McDermott, R. "Why Information Technology Internalization on Prosocial Behavior," Journal
Inspired but Cannot Deliver Knowledge Man of Applied Psychology (71:3), 1986, pp. 492
499.
agement," California Management Review
(41:4), 1999, pp. 103-117. Orlikowski, W. J. "Learning from Notes: Organi
McKeen, J. D., Guimaraes, T., and Wetherbe, J. zational Issues in Groupware Implementation,"
Information Society (9:3), 1993, pp. 237-251.
C. "The Relationship between User Partici
Osterloh, M., and Frey, B. S. "Motivation, Knowl
pation and User Satisfaction: An Investigation
edge Transfer, and Organizational Forms,"
of Four Contingency Factors," MIS Quarterly
Organization Science (11:5), 2000, pp. 538-550.
(18:4), 1994, pp. 427-451.
Patchen, M. Participation, Achievement, and
McKnight, D. H., Cummings, L. L., and Chervany, Involvement on the Job, Prentice Hall,
N. L. "Initial Trust Formation in New Organi Englewood Cliffs, NJ, 1970.
zational Relationships," Academy of Manage Pfeffer, J. Managing with Power: Politics and
ment Review (23:3), 1998, pp. 473-490. Influence in Organizations, Harvard Business
Microsoft. "Expanded MVP Program Puts Knowl School Press, Boston, 1992.
edge, Real-life Experience to Work for Microsoft Putnam, R. D. "The Prosperous Community:
Customers Worldwide," October 23, 2002 Social Capital and Public Life," The American
(available online at http://www.microsoft.com/ Prospect (4:13), 1993 (available online at
presspass/features/2002/Oct02/10-23mvp.asp). http://www.prospect.org/print/V4/13/putnam
Mintzberg, H. The Nature of Managerial Work, r.html).
Harper & Row, New York, 1973. Rheingold, H. The Virtual Community: Home
Mishra, A. K. "Organizational Responses to Crisis: steading on the Electronic Frontier, MIT Press,
The Centrality of Trust," in Trust in Organiza Cambridge, MA, 2000.
tions: Frontiers of Theory and Research,. R. M. Ruggles, R. "The State of the Notion: Knowledge
Kramer and T. R. Tyler (Eds.), Sage Publica Management in Practice," California Manage
tions, Thousand Oaks, CA, 1996, pp. 261-287. ment Review (40:3), 1998, pp. 80-89.
Mistzal, B. A. Trust in Modern Societies, Polity Ruppel, C. P., and Harrington, S. J. "Spreading
Press, Cambridge, England, 1996. Knowledge through Intranets: An Analysis of
Molm, L. D. Coercive Power in Social Exchange, the Organizational Culture Leading to Intranet
Cambridge University Press, NewYork, 1997. Adoption and Use," IEEE Transactions on
Moore, G. C, and Benbasat, I. "Development of Professional Communications (44:1), 2001, pp.
an Instrument to Measure the Perceptions of 37-52.
Adopting an Information Technology Innova Schultze, U., and Leidner, D. E. 'Studying Knowl
tion," Information Systems Research (2:3), edge Management in Information Systems
1991, pp. 173-191. Research: Discourses and Theoretical Assump
Nahapiet, J., and Ghoshal, S. "Social Capital, tions," MIS Quarterly (26:3), 2002, pp. 213-242.
Intellectual Capital, and Organizational Advan Sharma, S., Durand, R. M., and Gur Arie, O.
tage," Academy of Management Review (23:2), "Identification and Analysis of Moderator
1998, pp. 242-266. Variables," Journal of Marketing Research
Nunally, J. C. Psychometric Theory, McGraw-Hill, (18:3), 1981, pp. 291-300.
NewYork, 1978. Smith, D. H. "Altruism, Volunteers, and Volun
O'Dell, C, and Grayson, C. J. "If Only We Knew teerism," Journal of Voluntary Action Research
What We Know: Identification and Transfer of (10:1), 1981, pp. 21-36.

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Starbuck, W. H. "Learning by Knowledge-Inten About the Authors

sive Firms," Journal of Management Studies
(29:6), 1992, pp. 713-740 Atreyi Kankanhalli is an assistant professor in the
Stein, E. W., and Zwass, V. "Actualizing Organi Department of Information Systems at the National
zational Memory with Information Systems," University of Singapore (NUS). She obtained her
Information Systems Research (6:2), 1995, pp. B. Tech. from the Indian Institute of Technology
85-117. Delhi, M.S. from the Rensselaer Polytechnic
Stone, E. F. Research Methods in Organizational Institute, New York, and Ph.D. from NUS. She
Behavior, Goodyear, Santa Monica, CA, 1978. has been a Visiting Scholar at the University of
Stone, E. F., and Hollenbeck, J. R. "Some Issues California Berkeley. Professor Kankanhalli had
Associated with the Use of Moderated Regres considerable work experience in industrial R81D
sion," Organizational Behavior and Human and consulted for several organizations including
Performance (34), 1984, pp. 195-213. World Bank. Her work has appeared in the Journal
Thibaut, J. W., and Kelley, H. H. (1986). The of the American Society for Information Science
Social Psychology of Groups. New Brunswick, and Technology, Communications of the ACM,
Transaction Books. Decision Support Systems, International Journal of
Tsai, W., and Ghoshal, S. "Social Capital and Information Management, and the proceedings of
Value Creation," Academy of Management the International Conference on Information
Journal (41:4), 1998, pp. 464-476. Systems among others. She serves on several
Vallerand, R. J. "Toward a Hierarchical Model of information systems conference committees and
Intrinsic and Extrinsic Motivation," Advances in on the editorial board of the International Journal
Experimental Social Psychology (29), 1997, pp. of Knowledge Management. Professor Kankanhalli
271-360. has been awarded the Infocomm Development
Venkatesh, V., and Davis, F. D. "A Theoretical Authority Gold Medal at NUS and the ACM
Extension of the Technology Acceptance SIGMIS ICIS 2003 Best Doctoral Dissertation
Model: Four Longitudinal Field Studies," Man award. Her research interests are in knowledge
agement Science (46:2), 2000, pp. 186-204. management, virtual teams, electronic govern
Wasko, M. M., and Faraj, S. "It Is What One ment, and information systems security.
Does': Why People Participate and Help
Others in Electronic Communities of Practice," Bernard C. Y. Tan is an associate professor and
Journal of Strategic Information Systems Head of the Department of Information Systems at
(9:2/3), 2000, pp. 155-173. the National University of Singapore (NUS). He
Wasko, M. M., and Faraj, S. "Why Should I Share? received his Ph.D. in Information Systems from
Examining Knowledge Contribution in Networks NUS. Professor Tan has won several research
of Practice, "MIS Quarterly (29:1), 2005, pp. 35 and teaching awards at NUS. He has been a
57. Visiting Scholar in the Graduate School of Busi
Weill, P., and Olson, M. H. "An Assessment of the ness at Stanford University and the Terry College
Contingency Theory of Management Informa of Business at the University of Georgia. He is an
tion Systems," Journal of Management Infor elected Asia-Pacific Council Member for the
mation Systems (6:1), 1998, pp. 59-85. Association for Information Systems. Professor
Williamson, O. Markets and Hierarchies: Analysis Tan serves as a senior editor for MIS Quarterly
and Antitrust Implications, Free Press, New and e-Service Journal, and is on the editorial
York, 1975. boards of IEEE Transactions on Engineering
Yamagishi, T., and Cook, K. S. "Generalized Management, Journal of the AIS, Information and
Exchange and Social Dilemmas," Social Psy Management, Journal of Global Information Man
chology Quarterly (56:4), 1993, pp. 235-248. agement, and International Journal of Distance
Zack, M. H. "Managing Codified Knowledge," Education Technologies. His research has been
Sloan Management Review (40:4), 1999, pp. published in ACM Transactions on Computer
45-58. Human Interaction, ACM Transactions on Infor

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
mation Systems, Communications of the ACM, in Computer Science from the University of York
Decision Support Systems, European Journal of (United Kingdom). He is on leave from the
Information Systems, IEEE Transactions on National University of Singapore, where he is pro
Engineering Management, IEEE Transactions on fessor in the Department of Information Systems.
Professional Communication, IEEE Transactions He is past President of the Association for Infor
on Systems, Man, and Cybernetics, Information mation Systems. He has served on the editorial
and Management, Information Systems Research, boards of MIS Quarterly and Information Systems
International Journal of Human-Computer Studies, Research. His research has been published in
Journal of Management Information Systems, ACM Transactions on Computer-Human Interac
Journal of the AIS, Journal of the American tion, ACM Transactions on Information Systems,
Society for Information Science and Technology, European Journal of Information Systems, IEEE
Management Science and MIS Quarterly. His Transactions on Systems, Man, and Cybernetics,
current research focuses on cross-cultural issues, Information Systems Research, International
computer-mediated communication, knowledge Journal of Human-Computer Studies, Journal of
management, and information privacy. Management Information Systems, Journal of the
AIS, MIS Quarterly, and Management Science.
Kwok-Kee Wei is Chair Professor and Head of the His research focuses on computer-mediated com
Department of Information Systems at City munication, innovation and knowledge manage
University of Hong Kong. He received his D.Phil. ment, and human-computer interaction.

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Appendix
Survey Items ^ H
Construct Item Wording
Sharing my knowledge throu
lose my unique value in the or
lose my power base in the or
Sharing my knowledge throug
lose my knowledge that make
(LOKP) respect to othe
lose my knowledge that no on
I do not have the time to enter
EKRs (CEFF1) Or
It is laborious to codify my k
(CEFF2) Orliko
f The effort is high for me to
Effort int0 EKRs (C
?_._. I am worried that if I shar
* EKRs, I will have to spend
answering follow up que
I am afraid that my submission
additional clarifications or req
(CEFF5) (1998)
It is important to get a better wo
I share my knowledge throu
It is important to be promoted
knowledge through EK
Organizational It is important to ge
Reward my knowledge thro
(OREW) It is important to get a h
my knowledge through E
It is important to get more job s
my knowledge through EK
_|_j (1998)_

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Construct Item Wording and Code Source

Sharing my knowledge through EKRs improves my Adapted from Moore
image within the organization (IMAG1) and Benbasat (1991)
People in the organization who share their knowl- Adapted from Moore
edge through EKRs have more prestige than those and Benbasat (1991)
Imaoe who do not (IMAG2)
(\td\AC\ # Sharing my knowledge through EKRs improves Adapted from Green
( ' others recognition of me (IMAG3) (1989)
When I share my knowledge through EKRs, the Adapted from Kalman
people I work with respect me (IMAG4) (1999)
When I share my knowledge through EKRs, my Adapted from Kalman
superiors praise me (IMAG5) (1999)
When I share my knowledge through EKRs, I Developed based on
believe that I will get an answer for giving an answer Wasko and Faraj
(RECP1) (2000)
When I share my knowledge
_ . .A expect somebody to respo
(RECpT <RECP2> <1"3> knowledge to EKRs, I expect to Developed based on
When I contribute
get back knowledge when I need it (RECP3) feedback from sorters
When I share my knowledge through EKRs, I Developed based on
believe that my queries for knowledge will be feedback from sorters
answered in future (RECP4)
I have confidence in my ability to provide knowledge
that others in my organization consider valuable
(KSEF1)
-I have the expertise needed to provide valuable
f Fff knowledge for my organization (KSEF2) Adapted from Kalman
" It doesn't really make any difference whether I add (1999)
to the knowledge others are likely to share through
EKRs (KSEF3)
Most other employees can provide more valuable
knowledge than I can (KSEF4)
I enjoy sharing my knowledge with others through
EKRs(EHLPI)
I enjoy helping others by sharing my knowledge _
iiEnjoyment
Helping
in .,EKRs
i . _ through
Others
s ' _/,_ ,_,,.
(EHLP2)
u. ... _0/
, .. V .,. _ * a Developed based on
, u u Wasko and Faraj
^fhi p^ # 9 p someone e,se by shann9 my (200m
( ' knowledge through EKRs (EHLP3) ( }
Sharing my knowledge with others through EKRs
gives me pleasure (EHLP4)_

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00
Construct Item Wording and Code Source

I believe that people in my organization give credit
for other's knowledge where it is due (GTRU1)
I believe that people in my organization do not use
unauthorized knowledge (GTRU2) Developed based on
,rTRI I believe that people in my organization use other's Mishra (1996)
^ ' knowledge appropriately (GTRU3)
I believe that people in my organization share the
best knowledge that they have (GTRU4)
There is a norm of cooperation in my organization Developed based on
(PSNM1) Goodman and Darr
(1998)
There is a norm of collaboration in my organization Developed based on
(PSNM2) Goodman and Darr
Pro-Sharin (1998)
There is a norm of teamwork in my organization Developed based on
(PSNM) (PSNM3) Starbuck 1992
There is a willingness to value and respond to Develop
diversity in my organization (PSNM4) Leonard-Bar
There is a norm of openness to conflicting views in Develop
my organization (PSNM5) Leonard-Barton 1995
There is a norm of tolerance of mistakes in my Developed based on
organization (PSNM6) Leonard-Barton 1995
I am glad I chose to work for this organization rather
than another company (IDEN1)
I talk of this organization to my friends as a great
company to work for (IDEN2)
I am willing to put in a great deal of effort beyond
that normally expected to help my organization to
be successful (IDEN3)
I find that my values and my organization's values
Identification are very similar (IDEN4) Adapted from Cheney
(IDEN) In general the people employed by my organization 1983
are working toward the same goal (IDEN5)
I find it easy to identify myself with my organization
(IDEN6)
I feel that my organization cares about me (IDEN7)
I feel a sense of belonging towards my organization
(IDEN8)
I am proud to be an employee of this organization
(IDEN9)
What is your frequency of usage of. EKRs to Adapted from Igbaria et
contribute knowledge? (EUSG1) al. (1996)
EKR Usage I often use EKRs to contribute my knowledge in my Adapted from Davis
(EUSG) work. (EUSG2) (1989)
I regularly use EKRs to contribute my knowledge in Adapted from Davis
_ my work. (EUSG3) (1989)

112.207.190.2 on Fri, 31 May 2024 14:26:27 +00:00

Kankanhalli ContributingKnowledgeElectronic 2005

Uploaded by

Copyright:

Available Formats

You might also like

Kankanhalli ContributingKnowledgeElectronic 2005

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Kankanhalli ContributingKnowledgeElectronic 2005

Uploaded by

Copyright:

Available Formats

Contributing Knowledge to Electronic Knowledge Repositories: An Empirical

Stable URL: https://www.jstor.org/stable/25148670

Management Information Systems Research Center, University of Minnesota is collaborating

This content downloaded from

v Hid,! CwIlV secalssuE

Contributing Knowledge to Electronic

By: Atreyi Kankanhalli Abstract

1V. Sambamurthy and Mani Subramani were the

MIS Quarterly Vol. 29 No. 1, pp. 113-143/March 2005 113

This content downloaded from

114 MIS Quarterly Vol. 29 No. 1/March 2005

This content downloaded from

Theoretical Background 1HHH

MIS Quarterly Vol. 29 No. 1/March 2005 115

This content downloaded from

116 MIS Quarterly Vol. 29 No. 1/March 2005

This content downloaded from

MIS Quarterly Vol. 29 No. 1/March 2005 117

This content downloaded from

118 MIS Quarterly Vol. 29 No. 1/March 2005

This content downloaded from

Loss of Knowledge Power ^

Intrinsic Benefits ^-~ H6 _

Figure 1. The Research Model for EKR Usage by Knowledge Contr

MIS Quarterly Vol. 29 No. 1/March 20

This content downloaded from

120 MIS Quarterly Vol. 29 No. 1/March 2005

This content downloaded from

The American Productivity Quality Center website Image

MIS Quarterly Vol. 29 No. 1/March 2005 121

This content downloaded from

seemed to receive help more quickly when they

122 MIS Quarterly Vol. 29 No. 1/March 2005

This content downloaded from

Table 1. Formal Definitions of Constructs

Knowledge self- The confidence in one's ability to provide knowledge t

EKR usage (EUSG) The degree of EKR usage to contribute knowledge

Research Methodologywhere, new questions were developed ba

Table 1 provides formal definitions of the con

MIS Quarterly Vol. 29 No. 1/March 2005 123

This content downloaded from

Table 2. Results of Unstructured Sorting Exercise

124 MIS Quarterly Vol. 29 No

This content downloaded from

Table 3. Results of Structured Sorting Exercise

ment from each organization

MIS Quarterly Vol. 29 No. 1/March 2005 125

This content downloaded from

Table 4. Characteristics of Participating Organizations

126 MIS Quarterly Vol. 29 No. 1/March 2005

This content downloaded from

Table 5. Reliability of Constructs

Codification effort (CEFF) 5 0.85

Organizational reward (OREW) 4 0.96

Image (IMAG) 4 0.89

Enjoyment in helping others (EHLP) 4 0.96

Identification (IDEN) 6 0.96

EKR usage (EUSG) 3 0.85