PDF

Contributing Knowledge to Electronic Knowledge Repositories: An Empirical Investigation
Author(s): Atreyi Kankanhalli, Bernard C. Y. Tan and Kwok-Kee Wei

Source: MIS Quarterly, 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: http://www.jstor.org/stable/25148670 .
Accessed: 14/11/2014 15:15
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
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.
Management Information Systems Research Center, University of Minnesota is collaborating with JSTOR to
digitize, preserve and extend access to MIS Quarterly.
http://www.jstor.org
This content downloaded from 130.102.158.12 on Fri, 14 Nov 2014 15:15:49 PM

All use subject to JSTOR Terms and Conditions
Kankanhalli
et al./Contributing
Repositories
secalssuE
to Electronic
An Empirical
Knowledge
Contributing
Repositories:
Knowledge
Investigation1
Abstract
By: Atreyi Kankanhalli

Department of Information Systems
National University of Singapore
Singapore
Knowledge
CwIlV
Hid,!
3 Science
to Electronic
Organizations are attempting to leverage their

knowledge resources by employing knowledge
management (KM) systems, a key formof which
are electronic knowledge repositories (EKRs). A
Drive 2
117543
REPUBLICOF SINGAPORE
large number of KM initiatives fail due to the

reluctance of employees to share knowledge
through these systems. Motivated by such con
cerns, thisstudy formulatesand tests a theoretical
model to explain EKR usage by knowledge
contributors. The model employs social exchange
theoryto identifycost and benefit factorsaffecting
EKR usage, and social capital theory to account
atreyi@comp.nus.edu.sg
Bernard C. Y. Tan
National University of Singapore
3 Science Drive 2
Singapore 117543
REPUBLICOF SINGAPORE
for the moderating
btan@comp.nus.edu.sg
influence
of contextual
factors.
The model is validated through a large-scale

survey ofpublic sector organizations. The results
reveal thatknowledge self-efficacyand enjoyment
in helping others significantly impactEKR usage
Kwok-Kee Wei
City University of Hong Kong
83 Tat Chee Avenue
Kowloon
by knowledge contributors. Contextual factors

(generalized
HONGKONGSAR
trust, pro-sharing
norms,
and
identifi
cation) moderate the impact of codification effort,

reciprocity, and organizational reward on EKR
usage, respectively. Itcan be seen thatextrinsic
benefits (reciprocity and organizational reward)
impact EKR usage contingent on particular con
isweikk@cityu.edu.hk
textual factors whereas the effects of intrinsic

benefits (knowledge self-efficacyand enjoyment in
helping others) on EKR usage are notmoderated
and Mani
1V. Sambamurthy
Subramani
senior editors for this paper.
accepting
were
the
MIS
by contextual factors. The loss of knowledge

power and image do not appear to impact EKR
Quarterly
Vol. 29 No.
1, pp.
113-143/March

2005
113
Kankanhalli
et al./Contributing
to Electronic
Repositories
Knowledge
usage by knowledge contributors. Besides contri

buting to theorybuilding inKM, the results of this
study informKM practice.
Keywords: Knowledge management, electronic
knowledge repositories, knowledge contribution,
social exchange, social capital
Introduction
The
of organizational
strategic management
knowledge isa key factor thatcan help organiza
tions to sustain competitive advantage involatile
environments. Organizations are turningto knowl
edge management (KM) initiativesand techno
logies to leverage their knowledge resources.
According to the analyst firm IDC, business
spending on KM could rise from $2.7 billion in
2002 to $4.8 billion in 2007 (Babcock 2004).
Concurrent with the organizational interest inKM,
a large number of academic papers have been
published on KM (Schultze and Leidner 2002).
These developments reflectthesignificance ofKM
among
scholars
and
practitioners.
Knowledge management isdefined as "a systemic

and organizationally specified process for ac
quiring, organizing, and communicating both tacit
and explicit knowledge of employees so thatother
employees may make use of ittobe more effective
and productive in theirwork" (Alavi and Leidner
1999, p. 6). KM systems are "a class of informa
tion systems applied tomanaging organizational
knowledge. That is, they are IT-based systems
developed to support and enhance the organiza
tional processes of knowledge creation, storage/
retrieval, transfer, and application" (Alavi and
Leidner 2001, p. 114). Two models ofKM systems
have been identified in the informationsystems

literature: the repositorymodel and the network
The repository model
model (Alavi 2000).2
corresponds to the codification approach to KM
(Hansen et al. 1999). This approach emphasizes
codification and storage of knowledge so as to
labels these
(1999) alternately
integrative and interactive architectures
2Zack
114
two models
respectively.
as
facilitate knowledge reuse throughaccess to the

codified expertise. A key technological component
of this approach is electronic knowledge reposi
tories (EKRs) (Grover and Davenport 2001). The
networkmodel corresponds to the personalization
approach to KM (Hansen et al. 1999). This
approach emphasizes linkage among people for
the purpose of knowledge exchange.
Important
technological components of this approach are
knowledge directories that provide location of
expertise (Ruggles 1998) and electronic forum
software that allows people to interactwithin
communities of practice (Brownand Duguid 1991).
technological capabilities are important,

having sophisticated KM systems does not
inKM initiatives (Cross and
guarantee success
Baird 2000; McDermott 1999). This is because
social issues appear to be significant inensuring
knowledge sharing success (Ruppel and Harring
ton 2001). Both social and technical barriers to
usage of KM systems have been listedand calls
have been made to simultaneously address both
sets of issues (McDermott 1999; Zack 1999) in
order to be able to reap the benefits of KM that
have been experienced by some organizations
While
(Davenportetal.
1998; O'Dell and Grayson 1998).
This study focuses on EKRs since they are

fundamental to organizational knowledge capture
and dissemination, yet the factors affecting EKR
usage are not well understood (Markus 2001).
EKRs are electronic stores of content acquired
about all subjects forwhich the organization has
decided to maintain knowledge (Liebowitz and
Beckman 1998). EKRs can comprise multiple
knowledge bases as well as themechanisms for
acquisition, control, and publication of the knowl
edge.3 The process of knowledge sharing through
EKRs involves people contributingknowledge to
populate EKRs
(e.g., customer and supplier
knowledge, industrybest practices, and product
expertise) and people seeking knowledge from
of EKRs are
3According to the definition, the capabilities
to the mnemonic
functions of organizational
analogous
information systems
memory
1995).
(Stein and Zwass
of organizational
However,
memory
conceptualization
more on description at
information systems concentrates
the subsystem
level.
2005
MIS QuarterlyVol. 29 No. 1/March

Kankanhalli
EKRs for reuse. Success of EKRs requires that

knowledge contributorsbe willing to partwith their
knowledge and knowledge seekers be willing to
reuse the codified knowledge (Ba et al. 2001).
The distinctionbetween contributorsand seekers
is conceptual in that the same individualcan be a
contributoror a seeker at differentpoints in time.
from the
study examines EKR usage
as
this is
contributors
of
perspective
knowledge
This
the firststep toward knowledge leverage through

EKRs. Unless knowledge contributorsare willing
to provide content to EKRs, knowledge reuse
throughEKRs cannot take place.
Several prior studies have adopted a conceptual
(Kollock 1999; Markus 2001) or qualitative ap
proach (Goodman and Darr 1998; Orlikowski
1993; Wasko and Faraj 2000) in attempts to

understand the behavior of knowledge contribu
tors. Other studies have conducted experiments
(Constant et al. 1994) or surveys (Bock et al. 2005;
Constant et al. 1996; Jarvenpaa and Staples 2000;

and Faraj 2005) to model and explain
contributor behavior with varying success.
Wasko
Existing empirical studies focus on the benefits

(acting as motivators) ratherthan the costs (acting
as inhibitors)of knowledge contribution,despite
suggestions frompractitioner literature(O'Dell and

Grayson 1998) and conceptual literature(Ba et al.
2001) that cost factors can be important in
determining knowledge-sharing behavior.
study
advances
theoretical
This
development
on
knowledge contribution in two importantways.

First, itsimultaneously investigates both cost and
benefit factors affecting EKR usage. Second, it
incorporates contextual factors to illustratehow
these may moderate the relationships between
cost and benefit factors and EKR usage. The
results suggest organizational interventionsand
technologydesign considerations thatcan promote
knowledge contribution to EKRs, thereby faci
litatingreuse of organizational knowledge.
to Electronic
et al./Contributing
Knowledge
Repositories
edge contributors. In investigating the usage of

EKRs, the firstchoice of theoretical bases would
appear to be theories such as the technology
acceptance model (Davis 1989) which have been
successful inexplaining the usage of information
systems (e.g., Venkatesh and Davis 2000).
Although the technology acceptance model may

partially explain the behavior of knowledge
contributors,4thismodel does not directlyaccount
forthe social cost and benefit factorsexperienced
by knowledge contributors that may affect their

usage of collective technologies such as EKRs.
However, the social and individualcost and benefit
factors inknowledge sharing can be accounted for
by social exchange theory.The impactof cost and
benefit factors on EKR usage by knowledge
contributors is likely to be contingent upon
contextual factors (Constant et al. 1996; Goodman
and Darr 1998; Jarvenpaa and Staples 2000;

Orlikowski 1993). Social capital theoryaccounts
for several
important contextual factors in
knowledge exchange. Therefore, this study uses
the social exchange theoryand the social capital
theoryas itstheoretical bases.
Cost and Benefit Factors

Cost and benefit factors inour study are derived
based on social exchange theory. Social ex
change
theory
explains
human
behavior
in social
exchanges (Blau 1964), which differ from eco

nomic exchanges inthatobligations are not clearly
specified. Insuch exchanges, people do others a
favorwith a general expectation of some future
return but no clear expectation of exact future
return.
Therefore,
social
exchange
assumes
the
existence of relatively long-term relationships of

interestas opposed to one-off exchanges (Molm
1997). Knowledge sharing throughEKRs can be
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

dependent variable we are interested in
investigatingis thedegree of EKR usage by knowl
The
model has been suggested

4The technology acceptance
as a means
to extend the theoretical model developed
in
this study.

2005

115
Kankanhalli
et al./Contributing
to Electronic
Knowledge
Repositories
with the EKR serving as the intermediarybetween

knowledge contributorsand seekers. Knowledge
contributors share theirknowledge with no exact
expectation of future return. The quantity and

value of knowledge contributed is difficult to
specify and so is the returnobtained. Hence,
knowledge contributors are likelytowork on the
assumption of relatively longer-termrelationships

of interest.
Resources
(tangible and intangible) are the cur
rency of social
exchange.
Resources
given
away
during social exchange or negative outcomes of

exchange
can
be
seen
as
costs.
Resources
received as a resultof social exchange or positive

outcomes of exchange can be seen as benefits.
Social exchange theoryposits thatpeople behave

inways thatmaximize theirbenefits and minimize
theircosts (Molm 1997). Inagreement with this
that in
theory, researchers have suggested
creasing the benefits and reducing the costs for
contributing knowledge can help to encourage
knowledge sharing using KM systems (Markus

2001; Wasko and Faraj 2000), includingEKRs.
During social exchange, costs can be incurred in

the formof opportunity costs and actual loss of
resources (Molm 1997). Opportunity costs are
rewards foregone from alternative behavior not
chosen. For example, the timeand effortrequired

to codifyand inputknowledge intoEKRs (Ba et al.
2001; Markus 2001) can act as an opportunitycost
that precludes knowledge contributors from per
formingalternative tasks at that timeand accruing
the corresponding
rewards.
Also,
knowledge
con
tributors
may perceive a loss of power and unique
value within the organization associated with the
knowledge they transferto EKRs (Davenport and
Prusak 1998; Gray 2001). Such loss of knowledge
power can be considered as an actual loss of
resource during knowledge contribution.
During social exchange, benefits acting as moti
vators of human behavior can be extrinsic or
intrinsicinnature (Deci and Ryan 1980; Vallerand
Extrinsic benefits are sought after as
1997).
to ends desired by people. For example,
knowledge contributorsmay receive organizational
means
116
rewards for theircontributions (Beer and Nohria

2000; Hall 2001) throughwhich theycan obtain a
better lifestyle.As a resultof contribution,knowl
edge contributorsmay also enhance their image or
reputation in the organization (Ba et al. 2001;
Constant et al. 1994; Constant et al. 1996), which

can serve to increase their self-concept. By
sharing theirknowledge, contributorsmay receive

reciprocal benefits, i.e., their future requests for
knowledge being met by others (Connolly and
Thorn 1990; Kollock 1999; Wasko and Faraj
2000), which can facilitate theirwork. Intrinsic
benefits are sought after as ends by themselves.
For example, throughcontribution,knowledge con
tributorscan be satisfied by enhancing theirknowl
edge self-efficacyor confidence in theirability to

provide valuable knowledge that is useful to the
organization (Constant et al. 1994; Constant et al.
1996). Also, by contributingknowledge to EKRs,
knowledge contributors have the opportunity to
help others (Ba et al. 2001; Wasko and Faraj

2000). Previous studies on altruism have shown
that people enjoy and derive pleasure fromsuch
acts of helping others (Baumeister 1982; Krebs

1975). Research has established extrinsic and
intrinsicbenefits as motivators of human behavior
in several domains
(Vallerand 1997), including

knowledge sharing (Osterloh and Frey 2000).
Contextual
Factors
Contextual factors inour study are derived from

social capital theory. Social capital refers to the
resources embedded within networks of human
relationships (Nahapiet and Ghoshal
1998).
These networks include proximate as well as
virtual communities (Rheingold 2000).
Social
capital theoryposits thatsocial capital provides the

conditions necessary forknowledge exchange to
occur. Three key aspects of social capital thatcan
define the context for knowledge exchange are

trust, norms, and identification (Nahapiet and
Ghoshal 1998). Trust, norms, and identification
can be considered as social capital since theyare
organizational
resources
or assets
rooted
within
social relationships thatcan improve the efficiency
2005

Kankanhalli
of coordinated action. Practitioner literaturehas

the impacts of these factors without
considering whether their effects are direct or
described
However,
moderating.
several
prior
academic
studies (e.g., Constant et al. 1994; Jarvenpaa and

Staples 2000) have hinted at themoderating role
of these aspects of social capital in knowledge
sharing situations. Specifically, these three factors

are believed to be able to amplifyor dampen the
effects of particular cost and benefit factors on
knowledge-sharing behavior.
Trust is the belief that the intendedaction of others

be appropriate from our point of view
(Mistzal 1996). Itindicates a willingness of people
would
to be vulnerable to others due to beliefs in their
good
intent and
concern,
competence
and
capa
bility,and reliability(Mishra 1996). McKnight etal.

(1998) term these trustingbeliefs as benevolence
belief, competence belief, and predictabilitybelief,
respectively. Generalized trust is an impersonal
form of trust that does not rest with a specific
to Electronic
et al./Contributing
Knowledge
Repositories
A norm represents a degree of consensus in the

social system (Coleman 1990). Norms have the
effect of moderating human behavior in accor
dance with the expectations of the group or
community. Pro-sharing norms that have been
reported to enhance the climate for knowledge
sharing are norms of teamwork (Starbuck 1992),
collaboration and sharing (Goodman and Darr
1998; Jarvenpaa and Staples 2000; Orlikowski
1993), willingness to value and respond to

to conflicting views, and
diversity, openness
tolerance for failure (Leonard-Barton 1995). The
effects of costs and extrinsic benefits on knowl
edge contribution behavior are likelyto be con
tingenton these norms. The effects of intrinsic

benefits on knowledge contributionare not likelyto
be affected by contextual factors since these
benefits are seen as ends inthemselves. The cost
and extrinsic benefit factors may not influence

contribution behavior under conditions of strong
pro-sharing norms. Specifically,when pro-sharing
individualbut rests on behavior that isgeneralized
norms are strong, the costs of knowledge sharing

may not be a deterrent to knowledge contributors.
to a social
Conversely
unit as
a whole
(e.g.,
a community
of
workers
knowledge
knowledge
exchanging
throughEKRs) (Putnam 1993). In the context of
our study, generalized trust refers to the belief in
the good intent,competence, and reliabilityof
employees with respect tocontributingand reusing
knowledge through EKRs. With strong gener
alized trust,people may trusteach other without
having much personal knowledge about each
other. Generalized trusthas been viewed as a key
factor thatprovides a context forcooperation (Tsai
and Ghoshal
1998) and effective knowledge

exchange (Adler2001). When generalized trust is
strong, the effortrequired for knowledge sharing
may not be salient to knowledge contributors
because
theybelieve thatknowledge shared isnot

likelyto be misused by reusers (Davenport and
Prusak 1998). Conversely, when generalized trust
isweak, knowledge contributorsmay findtheeffort

required for knowledge sharing to be salient
because
they believe that others may inappro

use
theirknowledge. For example, con
priately
sultants at Ernst and Young declined tomake the
effort to contribute knowledge to repositories in
situationswhere trustdid not exist (Markus 2001).
when
pro-sharing
norms
are weak,
the
costs of contributionmay be salient. For example,

ithas been found that the costs of knowledge
sharing were a deterrent to KM system usage
there was
incompatibilitybetween the
collective nature of the technology and the com
petitive norm of the organizational context
(Orlikowski 1993). The extrinsic benefits of knowl
edge sharing may not be salient to knowledge
when
contributors
For
example,
when
pro-sharing
in environments
norms
where
are
strong.
such
norms
are strong, there is greater openness to the

potential for value creation through knowledge
exchange (Nahapiet and Ghoshal 1998) even
when extrinsic benefits to stimulate knowledge
contributionmay not exist.
Identificationisa conditionwhere the interestsof

with the interests of the
organization, resulting inthecreation of an identity
individuals merge
based on those interests (Johnson et al. 1999).

Identificationsets the context withinwhich com
munication and knowledge exchange occur among
organizational members (Nahapiet and Ghoshal

1998). Three components of identificationthat

2005

117
Kankanhalli
et al./Contributing
to Electronic
Knowledge
Repositories
have been identifiedinthe literatureare similarity

of values, membership in the organization, and
loyaltytoward the organization (Patchen 1970).
Similarity of values reflects the extent to which

members of an organization possess jointgoals
and interests.Membership is the degree towhich
self-concept ofmembers is linkedto the organiza
tion. Loyalty refers to the extent towhich mem
bers
support and defend the organization.

Identificationis likelyto provide a context forpro
social behavior by raising theconcern forcollective
interestswhich merge with the individual's own

interests (Johnson et al. 1999; O'Reilly and Chat
man 1986). Under conditions of strong identifica
tion, the effects of certain costs and benefits
pertaining to knowledge sharingmay be nullified in

the face of collective outcomes (Constant et al.
1996). Therefore,when identificationisstrong, the
effortrequired forknowledge sharing may not be

a deterrent to knowledge contributorsbecause the
concern fororganizational outcomes may domi

nate. Similarly, in such contexts, the need for
organizational reward forknowledge sharing may
not be salient to knowledge contributorsbecause
the regard forcollective outcomes
is strong.
contributors
factors
on
contingent
particular
contextual
norms,
trust, pro-sharing
(i.e., generalized
and identification).
Loss of Knowledge
Power
Previous research suggests thatby contributinga

part of their unique knowledge, knowledge con
tributors give up sole claim to the benefits
stemming from such knowledge (Gray 2001).

Therefore, knowledge contributors retain less
proprietaryknowledge upon which to argue their
value to the organization. This may reduce the
power position of knowledge contributors in
relation to the organization, making them more
replaceable. These arguments have also been
echoed by scholars in economics (Williamson
1975) and organization strategy (Mintzberg 1973;
Pfeffer 1992) in that uniqueness is considered a
key aspect of organizational power because the
lower the substitutability of an individual, the
greater is his or her power (Hickson et al. 1971).
The KM literaturereports the loss of power due to

knowledge contributionas a barrier to knowledge
sharing (Davenport and Prusak 1998; Orlikowski
1993). Since knowledge is perceived as a source
of power,
Research Model
and Hypotheses
The research model forexplaining EKR usage by
knowledge contributors incorporates constructs
from social exchange theory and social capital
theory (see Figure 1). Previous studies have

emphasized the importance of cost and benefit
factors indeterminingknowledge sharing behavior
(Ba et al. 2001; Markus 2001). Prior research has
highlighted the importance of contextual

social capital factors in influencingthe conditions
also
forknowledge sharing (Cohen and Prusak 2001;

Nahapiet and Ghoshal 1998). Therefore, cost
factors
cation
(i.e.,
effort)
loss of knowledge
and
benefit
power
factors
(i.e.,
and
codifi
organiza
tional reward, image, reciprocity,knowledge self

efficacy, and enjoyment in helping others) are
hypothesized to impactEKR usage by knowledge
118
knowledge
contributors
may
fear
losing
theirpower or value ifothers knowwhat theyknow

(Gray 2001; Thibaut and Kelley 1986). Potential
knowledge contributorsmay keep themselves out
of a knowledge exchange ifthey feel they can
benefit more by hoarding theirknowledge rather
than by sharing it(Davenport and Prusak 1998).
While the above discussion suggests a negative
relationshipbetween loss of knowledge power and
EKR usage by knowledge contributors, this rela
tionshipmay be contingent on pro-sharing norms.
such norms are strong, the barriers to

in contexts that
knowledge transferwitnessed
value personal expertise may be ineffective
(Jarvenpaa and Staples 2000). Insuch conditions
When
when
other
employees
are
seen
to be
sharing
knowledge, the deterrent effect of the loss of

knowledge power (which is relative to other em
ployees) may not be significant. Conversely with
weak pro-sharing norms, knowledge contributors
2005

to Electronic
et al./Contributing
Kankanhalli
Knowledge
Repositories
Costs
Loss of Knowledge
Power
Norms
Pro-sharing
Codification
"^^^^-^^^
^""^
I-:-1
Effort
H1
Generalized Trust
""^--^^^
H2a
Pro-sharing Norms
?,^^
H2b-___^^jj
-_
IdentificationH2c *
1-'
-^^^
""""??-._
Effort
l?I
^^^^^
CodificationEffort
Codification
^^^^^
I-1
Extrinsic Benefits
I-1
Organizational
Reward
Norms
Pro-sharing
N
?
_. w
H3a L
._.
Organizational Reward
Identification
-H3b_}.
'c
Image*
Pro-sharing Norms
I-1
Reciprocity
Pro-sharing Norms
[
T
Knowledge
Enjoyment
IntrinsicBenefits
o
_^
r
_-H4-"
- B
H5-
~"
^
^-~
Self-efficacy
inHelping
H6
^^.
H7
^-^"^
|_Others_|
Figure
1. The Research
Model
for EKR Usage
by Knowledge
Contributors

2005

119
Kankanhalli
et al./Contributing
to Electronic
Knowledge
Repositories
may be concerned about loss of knowledge power

when theycontribute knowledge to EKRs.
H2a: Codification effort is negatively related to

EKR usage by knowledge contributorsunder
conditions ofweak generalized trust
HI:
of knowledge power is negatively

related to EKR
usage
by knowledge
contributors under conditions of weak pro
Loss
sharing
norms.
Codification effortmay also be negatively related

to EKR usage by knowledge contributors contin
gent on pro-sharing norms. With strong teamwork
as well as cooperation and collaboration norms,

people tend not to be bothered about the effort
Codification Effort
The act of knowledge contribution to EKRs in
volves explication and codification of knowledge.
This can entail costs to knowledge contributorsas
an expense of time and effort (Ba et al. 2001;
Markus 2001). Efforthas been observed to be a
significant predictor of technology adoption

(Agarwal 2000). The time required forcodifying
knowledge can be considered as an opportunity
cost. Orlikowski (1993) reporteda situationwhere

consultants avoided knowledge contributiondue to
high opportunitycost. They were unwilling to use
the KM system as thiswould have required them

to incur non-chargeable hours or give up their
personal time.Aftercontributingknowledge, there
may be additional requests for clarification and

assistance fromknowledge recipients,which take
up more codification time from knowledge con
tributors(Goodman and Darr 1998).
The
above
arguments
suggest
a negative
relation
ship between codification effortand EKR usage by

knowledge contributors, but this relationship is
likelytobe contingent on generalized trust.Strong
generalized trust implies a general belief in the
intentof others (Putnam 1993). When

generalized trustisstrong, knowledge contributors
good
are
likely to believe that knowledge recipients

would notmisuse theirknowledge and would give
themcredit fortheircodification effort.Under such
circumstances, knowledge contributors may be
would be appreciated and

confident thattheireffort
may not be concerned about the effortthey have
to put inwhen contributing knowledge to EKRs.
Conversely, codification effortmay become a
deterrent to knowledge contributionwhen gener
alized trust isweak.
120
required tocontributeknowledge since others may

be likewise contributing. Conversely, when weak
pro-sharing norms prevail, knowledge contributors
tend to be aware of and hindered by the effort

required tocodifyknowledge forEKRs (Orlikowski
1993). Therefore, knowledge contributorsmay be

deterred by codification effortwhen pro-sharing
norms
are weak.
H2b: Codification effort is negatively related to

conditions
of weak
pro-sharing
norms.
Further, codification effort may be negatively

related to EKR usage by knowledge contributors
When identificationis
contingent on identification.
strong, collective concerns may dominate over

certain costs and benefits since collective interests
merge with the individual'sown interests (Johnson
et al. 1999). Insuch situations, knowledge contri
butors may not be concerned about theircodifi
cation effort inthe presence of the collective need
forcontributingknowledge to EKRs. Conversely,
when identificationisweak, codification effortmay

become a deterrent to knowledge contribution.
H2c: Codification effort is negatively related to

conditions ofweak identification.
Organizational Reward
To encourage EKR usage by knowledge contri
butors, organizations may provide various formsof
organizational reward such as increased pay,
bonuses, job security,or career advancement (Ba
et al. 2001; Beer and Nohria 2000; Hall 2001).
2005

Kankanhalli
The American ProductivityQuality Center website

provides examples of reward schemes for en
couraging knowledge sharing inorganizations (see

Several
http://www.apqc.org/).
com
consulting
panies have made knowledge sharing a basic

criterion for employee performance evaluation
(Davenport and Prusak 1998). There are also in
stances where employees enjoy better job security

as a resultof contributingknowledge (Hall 2001).
Although thisdiscussion suggests a positive rela
tionship between organizational reward and EKR
usage by knowledge contributors, the relationship
is likelyto be contingent on pro-sharing norms.
When norms of teamwork and collaboration are
strong, people may not require external incentives
toshare knowledge (Nahapiet and Ghoshal 1998).
Such a climate can remove the need fororgani

zational reward because knowledge contributors
may contribute their knowledge to EKRs
when
such
weak
pro-sharing
reward a
are
not available.
norms
can make
rewards
salient
motivator
even
organizational
for knowledge
H3a: Organizational reward ispositively related to

of weak
pro-sharing
norms.
Further, organizational reward may be positively

related to EKR usage by knowledge contributors
When identificationis
contingent on identification.
strong, the interests of employees and the
interestsof the organization merge (Johnson et al.
1999). Insuch situations, collective concerns tend
to
dominate
over
certain
costs
Knowledge
Repositories
Image
Inmost organizations today, the importance of
reputation is increasing as traditional contracts
between organizations and employees based on
lengthof service erode (Ba et al. 2001; Davenport
et al. 1998). Insuch working environments, knowl
edge contributorscan benefit fromshowing others
that they possess valuable expertise (Ba et al.
2001). This earns them respect (Constant et al.
1994) and a better image (Constant et al. 1996).
Therefore, knowledge contributors can benefit
from improvedself-concept when they contribute

their knowledge (Hall 2001; Kollock 1999).
Employees have been found to share theirbest

practice due to a desire to be recognized by their
peers as experts (O'Dell and Grayson 1998).
People who provided high-qualitytechnical knowl
edge have been found to enjoy better prestige in
theworkplace (Kollock 1999).
Conversely,
contribution.
conditions
to Electronic
et al./Contributing
and
benefits.
Therefore, under conditions of strong identification,

knowledge contributorsmay not require organi
zational reward to motivate them. Even when
organizational reward is absent, people may still
contributeknowledge toEKRs. Conversely, weak
identificationcan make organizational reward a
salient motivator forknowledge contributors.
H3b: Organizational reward ispositively related to

conditions of weak identification.
While thisdiscussion suggests a positive relation

ship between image and EKR usage by knowledge
contributors,the relationshipmay be contingenton
pro-sharing norms. When strong teamwork and
norms
collaboration
prevail,
contri
knowledge
butorsmay not requireextrinsic benefits (Nahapiet

and Ghoshal 1998) such as image in order to
contribute knowledge. Under such conditions,
knowledge contributorsare likelytocontribute their
knowledge to EKRs even ifbenefit in the formof
improved
sharing
image
norms
is absent.
can make
Conversely,
image
a salient
weak
pro
motivator
forknowledge contribution.
H4:
Image is positively related toEKR usage by

knowledge contributors under conditions of
weak
pro-sharing
norms.
Reciprocity
Reciprocity has been highlightedas a benefit for
individuals to engage in social exchange (Blau
1964). Itcan serve as a motivational mechanism
forpeople to contribute todiscretionary databases
(Connolly and Thorn 1990). Reciprocity can act as

2005

121
Kankanhalli
et al./Contributing
to Electronic
Knowledge
Repositories
a benefit forknowledge contributorsbecause
they
expect future help from others in lieu of their
contributions5 (Connolly and Thorn 1990; Kollock
1999). Prior research suggests that people who
share knowledge inonline communities believe in
Further,
reciprocity (Wasko and Faraj 2000).
researchers have observed that people who
regularly helped others in virtual communities
seemed to receive help more quicklywhen they
asked for it(Rheingold 2000).
These arguments suggest a positive relationship
between reciprocityand EKR usage by knowledge
contributors, but the relationshipmay be contin
gent on pro-sharing norms. When pro-sharing
norms are strong, knowledge contributors may
share their knowledge without need forextrinsic
benefits (Nahapiet and Ghoshal 1998) such as
reciprocity. Insuch a climate, knowledge contri
butors are likelyto contribute theirknowledge to

EKRs even in the absence of reciprocitybenefits.
Conversely,
when
pro-sharing
norms
are
weak,
reciprocitymay be a salient motivator for knowl

edge contributors.
H5:
Reciprocity is positively related to EKR

usage by knowledge contributors under
conditions of weak pro-sharing norms.
Knowledge
Self-Efficacy
Self-efficacy relates to the perception of people

about what theycan do with the skills theypossess
(Bandura 1986). When people share expertise
useful to the organization, theygain confidence in
terms of what they can do and this brings the
benefit of increased self-efficacy (Constant et al.
1994). This belief can serve as a self-motivational
force for knowledge contributors to contribute
knowledge to EKRs (Bock and Kim 2002; Kalman
of knowledge
refers to the expectation
Reciprocity
that their current contribution will lead to
contributors
This
is
their future request for knowledge
being met.
a
describe
different from pro-sharing
norms, which
climate to facilitate knowledge
sharing whereby sharing
for the costs and
may take place with littleconsideration
extrinsic benefits (including reciprocity) of doing so.
122
1999). Knowledge self-efficacy is typicallymani

fested in the formof people believing that their
knowledge can help to solve job-related problems
(Constant et al. 1996), improvework efficiency (Ba
et al. 2001), or make a difference to theirorgani
zation (Kollock 1999; Wasko and Faraj 2000).
Conversely, ifpeople feel that they lackknowledge
that isuseful to the organization, theymay decline
from contributing knowledge to EKRs because
they believe that theircontributioncannot make a
positive impact for the organization. The effectof

knowledge self-efficacyon EKR usage isnot likely
to be contingent on generalized trust,pro-sharing
or
norms,
H6:
identification.
Knowledge self-efficacy is positively related

toEKR usage by knowledge contributors.
Enjoyment inHelping Others

benefit is derived from the concept of
altruism. Altruism exists when people derive
intrinsicenjoyment from helping others without
expecting anything in return (Krebs 1975; Smith
This
1981). Although theremay be very few instances

of absolute altruism (involvingabsolute lackof self
concern
in the motivation
for an
act),
relative
altruism (where self-concern plays a minor role in

motivating an act) ismore prevalent (Smith 1981).
Knowledge contributorsmay be motivated by rela
tive altruism based on theirdesire to help others
(Davenport and Prusak 1998). Prior research
shows that knowledge contributors gain satis
faction by demonstrating theiraltruistic behavior
(Wasko and Faraj 2000). Such satisfaction stems

fromtheir intrinsic
enjoyment inhelping others (Ba
et al. 2001; Constant et al. 1994; Constant et al.
1996). Knowledge contributorswho derive enjoy
ment in helping others may be more inclined to
The effect of
contribute knowledge to EKRs.
on
EKR
in
others
usage is not
helping
enjoyment
likelyto be contingent on generalized
sharing
HI:
norms,
or
trust,pro
identification.
Enjoyment in helping others is positively

related to EKR
usage
by knowledge
contributors.
2005

Kankanhalli
1. Formal
Table
et al./Contributing
to Electronic
Knowledge
Repositories
of Constructs
Definitions
Construct
Definition
(Abbreviation)
Loss of knowledge
power (LOKP)
The perception of power and unique value lostdue to knowledge contributed

to EKR (Gray 2001)
Codification effort
The time and effortrequired to codify and inputknowledge intoEKR (Markus
2001)
(CEFF)
Organizational
reward (OREW)
The importanceof economic incentives provided forknowledge contribution

to EKR (Ba et al 2001; Hall 2001)
The perception of increase in reputationdue to contributingknowledge to
EKR (Constant et al 1996; Kollock 1999)
Image (IMAG)
Reciprocity (RECP)
The belief thatcurrent contribution to EKR would lead to future request for
knowledge being met (Davenport and Prusak 1998)
Knowledge selfefficacy (KSEF)
The confidence inone's ability to provide knowledge that isvaluable to the

organization via EKR (Constant et al. 1996; Kalman 1999)
Enjoyment inhelping
others (EHLP)
The perception of pleasure obtained fromhelping others throughknowledge

contributed to EKR (Wasko and Faraj 2000)
Generalized
trust
(GTRU)
Pro-sharing norms
(PSNM)
Identification(IDEN)
EKR usage (EUSG)
The belief inthe good intent,competence, and reliabilityof employees with

respect to contributingand reusing knowledge (Mishra 1996; Putnam 1993)
The prevalence of norms thatare intended to facilitateknowledge sharing in
the organization (Nahapiet and Ghoshal 1998; Orlikowski 1993)
The perception of similarityof values, membership, and loyalty
with the
al
et
Patchen
1999;
organization (Johnson
1970)
The degree of EKR usage to contribute knowledge (Davis 1989)
Research Methodology
The survey methodology was used to collect data
fortesting the research hypotheses. This method
chosen
because
it enhances
ology was
of
results
generalizability
(Dooley 2001).
Operationalization
Table
structs.
where, new questions were developed based on a

review of the previous KM and information
systems literature.One question forEKR usage
by knowledge contributors (the dependent con
struct)measured the frequency of EKR usage.6
The remaining50 questions inthe instrument
were
measured using seven-point scales anchored from
"stronglydisagree"
appendix).
of Constructs
to 'stronglyagree"
(see the
1 provides formal definitions of the con

Where
available,
these
constructs
were
measured using questions adapted from prior

studies to enhance validity (Stone 1978). Else
6We have to be mindful of the limitation that self-reported

EKR usage may be inflated.

2005

123
Kankanhalli
et al./Contributing
2. Results
Table
to Electronic
Knowledge
of Unstructured
Repositories
Sorting
Exercise
Actual Category
Hit
0-l_5_)CLl_,Q.3S-pOjK
Total
Rate
16
75
20
75
20
100
20
70
75
16
87.50
EHLP
1616
100
GTRU
1616
100
24
83.33
36
86.11
Target
__i_iyôiii-jo_zgc02
Category
3 S O I
LOKP
12
S S 5 5 2 Q 5 5
2 2
CEFF
15
OREW
20
2
IMAG
14
RECP
6
14
KSEF
PSNM
IDEN 1
20
31
Conceptual
Validation
that the questions

were
adapted
for measuring
from various
sources
the
IDEN5) were dropped because the sorters feltthat

these questions were ambiguous or did not fit in
with the other questions.
or
developed forthisstudy, all of the questions were

subjected to a two-stage conceptual validation
exercise based on procedures prescribed by
Moore and Benbasat (1991). Four graduate stu
dents participated in the firststage (unstructured
sorting) as sorters. Each sorterwas given the 51
questions printed on cards and mixed up. They
had to sort the questions by placing related ques
tions together and giving a label to each set of
related questions (which made up a construct).
This process helped to identifyambiguously
worded questions. The labels given by the four

sorters for the constructs corresponded very
closely to the names of the actual constructs.
Overall, the foursorters correctlyplaced more than
86 percent of the questions onto the intended
constructs (see Table 2). One question each for
image (IMAG5) and pro-sharing norms (PSNM3)
and two questions for identification(IDEN3 and
124
100
86.54
Average
constructs
(%)
12
EUSG 12
Given
Qs
Another fourgraduate students participated inthe

second stage (structuredsorting) as sorters. Each
sorter was given the 47 reworded questions
printed on cards and mixed up. Unlike the pre
vious
stage,
they were
given
the
names
and
definitionsof the constructs. They had to sort the

into a
questions by placing each question
construct category or an "other" (no fit)category.
Apart from two image questions (IMAG3 and
IMAG4) thatwere placed in the "other"category,
all sorters correctlyplaced all of the questions onto
the intended constructs (see Table 3). These two
questions were reworded based on suggestions
fromthe sorters. Two additional questions forreci

procity (RECP3 and RECP4) were added based
on suggestions from the sorters. Given that it is
desirable to have a minimum of three questions
per construct (Kim and Mueller 1981), these
additional questions helped to improvemeasure
2005

Kankanhalli
Table
3. Results
of Structured
Sorting
et al./Contributing
to Electronic
Repositories
Knowledge
Exercise
Actual Category
Category
3 g ?
LOKP
Hlt
Q-lL.3ti)Q-li_(L=>5.-e>5K
Target
16
û-wSow-ifczgcoS
S S 5 O ? 8 5 5
Total
Rate
Qs
(%)
16 100
20 20 100
CEFF
OREW
20
20
100
IMAG
14
16
87.5
RECP
100
KSEF
1616
100
EHLP
16
16
100
GTRU
1616
100
20
100
28
100
12
100
PSNM 20
IDEN 28
EUSG 12
98.86
Average
ment properties for reciprocity. All 49 questions

were then consolidated into an instrument for
survey
administration.
from each organization helped to identifyand

distribute the survey to colleagues who had used
EKRs as knowledge contributors in the course of
theirwork. Hence, all respondents to this survey

were KM practitioners familiarwith EKR usage
from
Survey Administration
The field studywas conducted inSingapore, over
a period of 6 months, frommid to late 2002.
Singapore isa developed countrywhere a number
of public organizations are in the process of
embarking on KM initiatives. Interviewswere con
ducted with key personnel incharge of KM initia

tives inthese organizations. Altogether, senior KM
executives from 17 public organizations were
interviewed.
Of the 17 public organizations contacted, 10

organizations (covering 7 industries)were willing
to participate in the survey. The organizations
were offered a report of our findings as an
incentive to participate. A senior KM executive
the perspective
of knowledge
contribution.
Among the 400 surveys distributed in these

organizations, 150 responses were obtained
yielding a response rate of 37.5 percent (see
Table 4). Completed surveys were returnedto the
authors by the respondents using envelopes with
pre-paid
postage.
Most of the 150 respondents were males (57.3%)

and inthe age group of 21 to 29 years (50.7%). A
majorityof the respondents had universitydegrees
(86.7%), with the rest having high school educa
tion. Their work covered functionalareas such as
informationsystems (33.3%), corporate services
(10%), marketing (8.7%), strategic planning

(8.7%), human resource (7.3%), and research and
development
(7.3%).
On
average,
the
respon
dents had work experience of 7.5 years and had

2005

125
et al./Contributing
Kankanhalli
Table
to Electronic
4. Characteristics
Knowledge
of Participating
Repositories
Organizations
Number of Responses
Industry
of Responses
Percentage
Defense
23 15.3%
11 7.3%
Education
Information
Technology
61 40.7%
6 4.0%
Library
Real Estate and Construction
17
Rehabilitation
11.3%
8 5.3%
24
Research and Development
16.0%
Total_150_100.0%_
been using EKRs as knowledge contributors for

2.5 years. They used informationtechnology
extensively in the course of theirwork.
the organizational mechanisms surrounding their

initiatives. These similarities allowed the
responses fromthe 10 organizations to be pooled
KM
for data
All organizations surveyed used EKRs as a critical

component of theirKM initiative.For instance one
organization in the education sector used a Lotus
Notes-based EKR to store student counseling
case studies and project reviews. Another organi
zation in the real estate sector used their Lotus
system for storing case studies,
Notes-based
reviews,
project
and
lessons
learned
(after action
reviews). All EKRs studied were mainly used to
store
reviews,
project
case
lessons
studies,
learned, and best practices. The contents of the

EKRs were inthe formof documents (Word or pdf
format) and presentations (PowerPoint format).
Keywords and othermetadata were used to index
and retrieve the content. Therefore, the contents
were fairlystructured. The number of EKR users
in these organizations varied fromabout 120 to

2,000, while the organization sizes ranged from
about 250 to2,500. All organizations provided KM
training and incentives for knowledge sharing.
There was also topmanagement support forKM
initiatives.
However,
dates foremployees
systems),
i.e.,
EKR
there were
no explicit
to use EKRs
usage
was
man
(or other KM
voluntary.
Informationobtained and observations of EKRs

during interviews revealed that all of the 10
organizations participating in this survey were
comparable intermsof theirEKR capabilities and
126
analysis.
Data Analysis
and Results
for reliability
The constructs were firstassessed
and validity.Afterascertaining that the constructs
could
meet
parametric
requirements
of the regres
sion test, the hypotheses were tested using

moderated multiple regression analysis. All statis
tical testswere carried out at a 5 percent level of
significance.
Reliability and Validity

The constructs were assessed
for reliabilityusing
(Cronbach 1951). Nunnally
(1978) suggested that a value of at least 0.70
indicated adequate reliability. Inorder to improve
the reliabilities of the corresponding constructs,
one question was omitted from each of the
reward
following constructs:
organizational
(OREW1), reciprocity(RECP1), pro-sharing norms
Cronbach's
alpha
Subse
(PSNM5), and identification (IDEN6).
quently, all of the constructs had adequate
reliability(see Table 5).
2005

Kankanhalli
Table
5. Reliability
Number of Questions
3 0.85
4 0.96
Knowledge self-efficacy (KSEF)
Enjoyment inhelping others (EHLP)
Identification(IDEN)
(EUSG)
0.96
4 0.85
trust (GTRU)
Pro-sharing norms (PSNM)
4 0.93
6 0.96
3 0.85
The questions were tested forvalidityusing factor

analysis with principal components analysis and
Convergent
validity
was
as
sessed by checking loadings to see ifitemswithin

the same construct correlate highly amongst
themselves.
Alpha
4 0.89
Reciprocity (RECP)
rotation.
Cronbach's
4 0.96
Organizational reward (OREW)
varimax
Repositories
5 0.85
Codification effort(CEFF)
EKR usage
Knowledge
4 0.95
Loss of knowledge power (LOKP)
Generalized
to Electronic
of Constructs
Construct
Image (IMAG)
et al./Contributing
Discriminant
validity
was
assessed
by examining the factor loadings to see ifques

tions loaded more highly on their intended
constructs than on other constructs (Cook and
Campbell 1979). Loadings of 0.45 to 0.54 are
considered fair,0.55 to 0.62 are considered good,
0.63 to 0.70 are considered very good, and above
0.71 are considered excellent (Comrey 1973).
Factor analysis yielded 11 components with
eigenvalues above 1 (see Table 6). These 11
components corresponded to the 11 constructs.
Two questions forcodification effort (CEFF4 and
CEFF5) tapped onto other constructs and were
omitted. All other questions had at least good
loadings on their intended constructs. After
omitting the two questions, the reliabilityof the
codification effortconstruct improved to 0.91.
Hypotheses
Tests
Studies in information
systems (e.g., McKeen etal.
1994; Weill and Olson 1989) and inother disci
plines (e.g., Jehn et al. 1999) have used moder
ated multiple regression to test interactioneffects.
Moderated multiple regression is a hierarchical
procedure thatfirsttests the relationship between
independent constructs and the dependent con
struct, and then tests the relationship between
interaction terms and the dependent construct
(Sharma et al. 1981; Stone and Hollenbeck 1984).
Interactiontermsare computed bymultiplyingtwo
independent constructs. A significant change in
explanatory power between the two steps, which
can be assessed by lookingat the significance of
the change inF value, indicates the presence of
moderating effects.
In this study, the independent constructs were

entered in the firststep of regression and the
interactiontermswere added in the second step.
All
interaction
terms
were
assessed
simulta

2005

127
et al./Contributing
Kankanhalli
to Electronic
Knowledge
Repositories
6. Validity of Questions
I Table
Question
LOKP1
LOKP2
LOKP3
LOKP4_-0.16
CEFF1
CEFF2
CEFF3
CEFF4
CEFF5_-0.06
_^___^_Factor_
1|2|3|4|5|6|7|8|9|10|11
-0.08
-0.20
0.03 -0.15
0.01
-0.04
0.00
-0.16
-0.13
-0.23
-0.17
0.00
0.01
-0.13
-0.08
0.87
0.90
0.90
-0.23
0.09
-0.10
Q.87
0-03
-0.13
-0.19
-0.14
-0.08
-0.15
0.06
-0.14 -0.08
0.00 -0.07
-0.10 -0.13
0.01
-0.08
-0.11
-0.09
0.09
0.10
0.04
0.14
-0.05
0.00
0.04
-0.07
-0.06
-0.17
-0.15
-0.12
-0.09
-0.12
0.16
-0.09
-0.11
OREW2
OREW3
OREW4
0.03
0.07
0.07
OREW5_0.07
IMAG1
IMAG2
IMAG3
0.03
0.00
0.05
0.03
0.13
0.03
0.12
IMAG4_0.07
RECP2
RECP3
0.19
-0.01
-0.15
-0.06
0.04
0.00
0.03
0.00
-0.04
-0.09
-0.11
-0.05
-0.04
-0.11
-0.09
-0.02
0.00
-0.06
0.00
0.85
0.89
0.90
0.40
-0.19
-0.04
-0.02
-0.15
-0.09
-0.04
-0.08
-0.42
Q.39
-0.10
-0.13
-0.43
0.11
0.06
0.08
-0-05
0.01
0.06
0.08
-0.01
0.28
0.24
0.20
0.13
0.04
0.06
0.04
0.01
0.01
-0.02
-0.01
0.10
0.09
0.12
0.07
0.04
0.10
0.11
0.11
0.11
0.14
0.09
0.11
0.01
0.21
0.03
0.22
0.15
0.88
0.91
0.92
Q.89
0.37
0.19
0.26
0.24
-0.08
0.00
0.04
0.01
-0.05
0.18
-0.07
-0.08
0.09
0.03
0.13
0-04
0.03
-0.01
-0.02
0.02
-0.11
0.25
-0.09
-0.04
0.18
0.10
0.18
0.13
0.07
0.05
0.04
0.07
0.16
0.09
0.01
0.21
0.19
0.09
0.11
0.13
0.21
0.13
0.01
0.01
0.07
0.05
0.10
0.03
0.07
-0.08
0.02
0.01
-0.11
-0.11
-0.13
-0-11
0.69
0.82
0.78
Q.86
0.10
0.21
0.18
-0.04
0.03
0.04
-0-03
-0.03
0.02
-0.02
0.10
0.11
0.08
0.06
0.15
0.03
0.11
0.04
0.07
0.04
0.04
0.33
0.01
0.07
0.02
0.08
0.05
0.03
0.71
0.87
Q.88
0.00
0.07
0.09
0.04
0.15
0.02
0.04
0.09
0.15
0.12
0.11
-0.24
-0.27
-0.26
0.16
0.14
0.15
0.03
0.04
0.04
0.18
0.11
0.04
0.08
0.12
0.12
0.11
0.08
0.03
0.15
0.12
0.08
0.15
0.08
0.02
0.07
0.05
0.08
0.25
0.18
0.19
0.37
0.18
0.12
0.09
0.17
0.08
0.10
0.20
0-24
0.09
0.58
0.17
0.74
0.08
0.79
0-02
Q.63
0.24 -0.07
0.19 -0.05
0.05
0.25
0.25
0.08
0.11
-0.06
0.04
0.02
0.11
0.11
0.07
0.13
0.13
0.01
0.07
0.13
0.15
0.16
0.27
0.15
0.09
0.07
0.04
0.06
0.02
0.04
0.12
0.08
-0.03
-0.04
0.02
0.10
0.82
0.63
0.62
1.67
3.88
0.16
0.10
0.06
KSEF4_0.03
EHLP1
EHLP2
EHLP3
0.15
0.18
0.22
0.82 -0.02
0.86 -0.02
0.84 0.08
Q.81
0.08
0.24
-0.27
GTRU1
GTRU2
GTRU3
0.37
0.23
0.29
GTRU4_0.31
PSNM1
PSNM2
PSNM3
0.36
0.35
0.42
0.00
0.16
0.19
0.01
0.04
0.07
0.09
0.01
0.20
0.09
0.07
0.05
0.03
0.05
0.03
0.09
0.05
0.00
0.10
0.09
0.14
0.18
0.08
0.07
-0.11
-0.15
-0.06
-0.08
-0.08
-0.06
0.01
0.04
-0.01
-0.02
0.08
-0.05
0.11
0.10
0.06
0.09
0.80
0.86
0.84
0.87
0.82
0.85
Ô04~
0.18
0.16 0.10
0.16 0.13
0.15 0.11
0.16 0.03
0.04 -0.01
0.10 -0.06
0.02
0.05
0.05
0.07
0.06
0.17
-0.15
-0.13
-0.08
-0.14
-0.09
-0.08
0.06
0.03
0.09
0.05
0.07
0.06
0.81
0.82
0.70
Q.66
0.22
0.16
0.19
0.19
0.18
0.15
-0.03 ~O06"
0.25
0.38
0.42 0.21
3.98
3.85
13.16
9.27 8.96
0.23
0.19
0.20
3.76
8.75
-0.04 ~oTj
0.07
-0.06
-0.10
0.05
2.97
3.67
6.92
8.54
013"
0.04
0.00
2.87
6.68
0.10
0.09
0.08
2.69
6.27
0.05
0.03
-0.04
2.66
6.19
-0.05
0.23
0.17
2.41
5.59
13.16
22.43 31.39
62.28
68.55
74.74
80.33
PSNM4_0.43
IDEN1
IDEN2
IDEN3
IDEN4
IDEN5
IDEN7
EUSG1
EUSG2
EUSG3_0.21
Eigenvalue_5.70
Variance
explained
(%)_
Cumulative
variance
128
-0.03
-0.15
-0.10
-0.12
-0.11
RECP4_0.10
KSEF1
KSEF2
KSEF3
EHLP4_0-16
-0.08
0.00
-0.04
0.87
0.89
0.92
Q.91
0.16
0.18
0.14
40.14
48.68
55.60
(%)
2005

0.01
0.03
0.09
0.15
84.21
Kankanhalli
Table
7. Results
of Hypothesis
et al./Contributing
to Electronic
Knowledge
Repositories
Tests
Standardized
Coefficient
Hypothesis Test
Step 1: Main Effects

Loss of knowledge power (LOKP) 0.05
Codification effort(CEFF)
-0.07
reward
Organizational
(OREW) 0.22**
-0.05
Image (IMAG)
(RECP) 0.11
self-efficacy (KSEF)
inhelping others (EHLP)
trust (GTRU) -0.13
norms
Pro-sharing
(PSNM) 0.04
Identification(IDEN) 0.04
Reciprocity
Knowledge
Enjoyment
Generalized
R2
Adjusted R2
H6 was supported
H7 was supported
0.25***
0.43***
0.44
0.38
F 10.95***
Step 2:
Interaction Terms
LOKP^PSNM
-0.11
CEFF-GTRU
-0.18*
CEFF-PSNM
0.01
CEFFHDEN
0.07
OREW-PSNM
-0.17
OREWHDEN
0.23**
IMAG'PSNM
0.12
RECP?PSNM
-0.18*
R2
H1 was not supported

H2a was supported
H2b was not supported
H2c was not supported
H3a was not supported
H3b was not supported
H4 was not supported
H5 was supported
0.52
Adjusted R2
0.45
F 7.72***
R2 change
0.08
F change
*p
<
0.05;
3.23**
**p
<
0.01,
***p
< 0.001
neously so that theireffects could be seen in the

context of the overall model (i.e., inthe presence
of othermain and interactioneffects). To alleviate

possible collinearity problems, the values of all
constructs were centered (mean subtracted)
during regression (Aiken and West 1991). The R2
value of 0.52 and adjusted R2 value of 0.45 (F =
7.72, p < 0.001) indicated that the overall model
was more than satisfactory in explaining the
variance in EKR usage by knowledge contri
butors.7 The change inR2 value between the two

=
steps of regression was 0.08 (change inF 3.23,
<
p
0.01), indicating that the outcome of the
second step (i.e., testing of interaction terms)
could be interpreted(see Table 7).
7Falkand Miller (1992) indicatethatexplanatorypower

(R2 value)
greater
than 10 percent
is acceptable.

2005

129
Kankanhalli
to Electronic
et al./Contributing
Knowledge
Repositories
Table 7 summarizes the results of hypotheses

tests. Loss of knowledge power had no significant
relationshipwith EKR usage even under conditions
of weak
pro-sharing
norms,
i.e., H1 was
not sup
ported. Codification efforthad a significantnega

tive relationshipwith EKR usage under conditions
ofweak generalized trustbut not under conditions
ofweak pro-sharing norms and weak identification,
i.e.,H2a was supported but H2b and H2c were not
supported. Organizational reward had a significant
positive relationshipwith EKR usage, not under
conditions of weak pro-sharing norms but under
conditions of strong identification(thiswas oppo
site to the prediction of hypothesis H3b), i.e.,H3a
and H3b were not supported.
Image had no
significantrelationshipwith EKR usage even under

conditions ofweak pro-sharing norms. Hence, H4
was not supported. Reciprocity had a significant
positive relationshipwith EKR usage under con
ditions
of weak
pro-sharing
norms,
i.e., H5 was
supported. Knowledge self-efficacyand enjoyment

inhelping others had significantpositive relation
ships with EKR usage, i.e., H6 and H7 were
supported.
The standardized coefficients inTable 7 indicate

that,relatively,enjoyment inhelping others had the
strongest impact on EKR usage by knowledge
contributors followed by knowledge self-efficacy

and organizational rewards (both directly and
moderated by identification). Codification effort
moderated by generalized trust and reciprocity
moderated by pro-sharing norms had the least
impactamong the significantdeterminants of EKR
usage by knowledge contributors.
Control Variables
Furtheranalysis was carried out tomake sure the
significant resultswere not due to covariation with
controlvariables. Previous literaturesuggests that
gender (Jarvenpaa and Staples 2000), age

(Jarvenpaa and Staples 2000), work experience
(Constant et al. 1994), and education (Constant et
al. 1994) may affect knowledge contribution
behavior. EKR user community size may also
influence EKR usage. These control variables
(age,
130
gender,
education,
work
experience,
and
community size) were included in a moderated

multiple regression model together with the 11
original constructs. The results demonstrated that
the significantmain effects and interactionterms
remained the same as inTable 7. None of the
control variables had a significant impacton EKR
usage. Also, the inclusionof the control variables
did not significantly increase the variance ex
plained. Therefore, the resultsof hypotheses tests
to be stable and
(see Table 7) appeared
independent of control variables.
Discussion
and Implications WM
Based on our findings, loss of knowledge power

did not significantlyaffect EKR usage by knowl
edge contributors, not even under conditions of
weak pro-sharing norms. This may be due to the
fact thatknowledge contribution isvoluntary inthe
organizations surveyed in thisstudy. Under such
circumstances, knowledge contributorscan decide
what to contribute toEKRs. Hence, theyneed not
fear thattheirknowledge contributionwould render
them less valuable to the organization. The
respondents of this survey are actively partici
pating inthe creation or acquisition of new knowl
edge
remain
of their work.
in the course
valuable
to
the
Thus,
organization
they can
even
after
contributing knowledge to EKRs. Another plau

sible explanation for this result is self-selection
among
the survey
respondents.
Knowledge
contri
butorswho responded to this survey may also be

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.
The relationship between codification effortand

EKR usage by knowledge contributors was
contingent on generalized trust.As hypothesized,
when generalized trust isstrong, codification effort
may not be a deterrent forEKR usage by knowl
edge contributors. However, the relationship
between codification effort and EKR usage by
knowledge contributorswas not contingenton pro
sharing norms and identification.Therefore, even
when there are norms of collaboration and
cooperation or when the interests of people are
2005

Kankanhalli
aligned with those of the organization, such

contextual factors do not impact the relationship
between codification effortand EKR usage by
knowledge contributors.The existing literature(Ba
et al. 2001; Goodman and Darr 1998) has dis
cussed the deterrenteffectof codification efforton
knowledge contribution.The findingsof thisstudy
extend the previous literatureby revealing that the
relationship between codification effort and
knowledge contribution to EKRs is salient when
generalized trust isweak.
The relationship between organizational reward
and EKR usage by knowledge contributorswas
both direct and contingent on identification.
Contrary to hypothesis H3b, this relationship
appears to be stronger when identification is
strong, i.e.,when knowledge contributors to EKR
share the same interestsas theorganization, they
tend to be more motivated by organizational
rewards. Itappears that ifknowledge contributors
do not share the interestsof theorganization, even
organizational reward may not motivate them to
contribute theirknowledge to EKRs. The relation
ship between organizational reward and EKR
usage
was
not contingent
on
pro-sharing
to EKRs
even
after pro-sharing
norms
have developed. The findingsof thisstudy extend

prior literatureby revealing that the relationship
between organizational rewardand EKR usage by
knowledge contributors ismost significantwhen
identificationis strong.
Image did not significantlyaffect EKR usage by
knowledge contributors,noteven under conditions
ofweak pro-sharing norms. This may be due to
the dual effects of pro-sharing norms. On one
hand, strong teamwork and collaboration norms
may reduce the need for improved image as a
motivator forknowledge contributiontoEKRs. On
theother hand, strong error tolerance and diversity
normsmay enhance the need forthisbenefitsince
the riskofmaking mistakes during contributionare
lessened.
The
converse
dual
effects
may
when
to Electronic
norms
pro-sharing
Repositories
Knowledge
are weak.
Prior
literature
suggests that increased recognitionby colleagues

or the organizational community can be an impor
tantmotivator for employees to contribute their
knowledge (Constant et al. 1994; Hall 2001;
Kollock 1999; O'Dell and Grayson 1998). How
ever, ourfindings suggest thatfutureresearch may
want to furtherexamine the role of pro-sharing
norms in relation to image as a motivator for
knowledge contribution.
relationship between reciprocityand EKR
usage by knowledge contributorswas contingent
on pro-sharing norms. As hypothesized, when
The
pro-sharing
norms
are
strong
and
there
is a
climate of collaboration and cooperation, knowl

when
edge contributorsdo not look for reciprocity
to
their
EKRs.
However,
contributing
knowledge
when
pro-sharing
norms
are
weak,
reciprocity
benefit isa motivator forknowledge contributionto

EKRs.
This finding extends prior literature
(Connolly and Thorn 1990; Kollock 1999) by
indicatingthat the relationshipbetween reciprocity
and EKR usage by knowledge contributors is
significant
when
pro-sharing
norms
are weak.
norms.
Several organizations have used organizational

reward (Ba et al. 2001; Beer and Nohria 2000; Hall
2001) to build up pro-sharing norms among their
Being used to obtaining organiza
employees.
tional reward, knowledge contributors may
continue to expect such rewards for knowledge
contribution
et al./Contributing
occur
Knowledge self-efficacysignificantlyimpactedEKR
usage by knowledge contributors. As hypothe
sized, when people are confident of theirability to
contribute knowledge thatwould be useful to the
organization, they tend to be more motivated to do
so through EKRs. This result is consistent with
previous KM experiments (Constant et al. 1996)
and conceptual articles (Ba et al. 2001). Enjoy
ment inhelping others also significantlyaffected
EKR usage by knowledge contributors.As hypoth
esized, when people feel good about contributing
knowledge to help others, they tend to be more
motivated to do so through EKRs. Again, this
result is consistent with previous KM conceptual
(Ba et al. 2001) and case study literature
(Davenport and Prusak 1998) highlightingaltruism
as a motivator forknowledge sharing.
Implications for Theory

This study advances theoretical development in
the area of KM ingeneral and EKRs inparticular.

2005

131
Kankanhalli
??^
et al./Contributing
to Electronic
Knowledge
Repositories
Costs
Codification effort
Generalized Trust
Extrinsic
Benefits
Organizational
Organizational
\^^
\^
Reward-
^^^l-1
Reward
Identification^ BY
Reciprocity
Norms
Pro-sharing
-?-18*
i-:-1
n 22**
_0
^^
IntrinsicBenefits
Knowledge Self-efficacy
Enjoyment inHelping
Others
23**
~* KNOWLEDGE
I-:-1 ma*-~
"u-'?
>T
__^?
EKR USAGE
-^
I-1^^
^^ I-1
CONTRIBUTORS
-
0.25***
s'
R2 = 0.52
/^
^0.43***
'
^????^?_^_^^____^______________________________________________________
Figure
2. The Revised
Research
Model
It demonstrates that cost and benefit factors

derived fromsocial exchange theorymoderated by
contextual factors derived from social capital
theorycan predict EKR usage by knowledge con
tributors. Based on the findingsof this study,we
refineour original research model (see Figure 1)
and propose an alternative research model that

can better account forEKR usage by knowledge
contributors (see Figure 2). The explanatory
power of the revised research model demonstrates
the value of using social exchange theory and
social capital theory to predict usage of collective
technologies such as EKRs.
The resultsof thisstudy shed lighton how extrinsic
or intrinsicbenefits may differ in terms of their
impacton EKR usage by knowledge contributors.
132
The impacts of extrinsic benefits (organizational

rewards and reciprocity)appear to be moderated
by contextual factors (identification and pro
sharing norms, respectively). This indicates that
the provision of extrinsic benefits alone may not be
adequate motivators of knowledge contribution to
EKRs unless these extrinsic benefits are provided
in appropriate
contexts.
For
example,
organiza
tional reward seems to work best when

fication is strong while reciprocityseems
most
effective
when
pro-sharing
norms
identi
to be
are weak.
However the effects of intrinsicbenefits (knowl

edge self-efficacyand enjoyment inhelping others)
appear to be direct. Given that intrinsicbenefits
are sought as ends desired by people, contextual
factors do not play a significant role in influencing
the value of these benefits to knowledge contri
2005

Kankanhalli
butors. The fostering of intrinsicbenefits alone

may be sufficient to motivate knowledge contri
butors to contribute theirknowledge to EKRs, in
many contexts (differentcombinations of con
textual factors). The impactof cost factors (codifi
cation effort) appears

to be moderated by
contextual factors (generalized trust).This implies
thatmeasures to alleviate the costs of knowledge
contribution to EKRs may only be necessary in
may be useful to
specific contexts. For example, it
reduce codification effort

when generalized trust is
weak but itmay not be necessary to do thiswhen
generalized trust is strong. In summary, three
contextual factors that have been found to be
important are generalized
trust, pro-sharing
norms,
and identification.
factors thatdetermine EKR
Apart from identifying
usage by knowledge contributors, this study also
contributes to theory by unveiling factors that do
not appear to impact EKR usage by knowledge
contributors. For example, the loss of knowledge
power is thought to be a barrier to knowledge
contribution. However, inour study, this is not a
significant concern forknowledge contributors to

EKR. As another example, image isconsidered as
a motivator forknowledge contribution. However,
our findingsshow that knowledge contributors to
EKR may not be concerned about the image
associated with knowledge contribution. These
results suggest that futureresearch should take a
closer
look at how
power
and
image
are
perceived
the
by knowledge contributors. Besides identifying
constructs that can or cannot predict EKR usage
by knowledge contributors, this study also under
takes a rigorousconceptual and empirical process
to develop
measures
for each
of these
constructs.
et al./Contributing
gestions to management about how to promote

EKR usage by knowledge contributors. First,
management
can
raise
the perceptions
of knowl
Repositories
the self-efficacy of these knowledge
enhance
contributors.
Second, management can attempt to raise the

level of enjoyment that knowledge contributors
experience as theyhelp others. This may be done
by connecting knowledge contributorsand knowl

edge recipients in order to allow recipients to
express their appreciation for the knowledge
received. The realization that their colleagues
have benefitted fromtheirknowledge contribution
can increase the feeling of altruism among
knowledge contributors (Davenport and Prusak
1998). As a way of motivating knowledge contri
butors, knowledge seekers can be rewarded for
findingsolutions fromEKRs and acknowledging

the sources of the solutions. The Most Valuable
Professionals Program at MicrosoftCorporation is
an example of an initiativethat raises altruism (and
community spirit) in thisway. Through this pro

gram, people who have provided useful technical
assistance to other users ofMicrosoft technology
are identifiedand informedthat they have helped
others (Microsoft2002).
Third, organizational reward (such as betterwork

incentive,
promotion
assignment,
incentive,
salary
bonus incentive, or job security) seems to be

effective for encouraging EKR usage by knowl
edge
Collectively, the results of this study indicate the

circumstances under which organizational mea
sures to promote knowledge contributiontoEKRs
may be more effective. These results offersug
Knowledge
edge self-efficacy among valued knowledge

contributors by indicating to them that their
knowledge contributionmakes a significant dif
ference to the organization. This can be done by
highlighting the improved organizational perfor
mance arising fromtheirknowledge contributions.
Organizations such as Amazon.com
regularly
recognize theirtop reviewers, serving as a way to
contributors.
various
Implications forPractice
to Electronic
forms of
Organizations
have
used
reward
to
organizational
to
contribute
their
knowl
encourage employees
edge toEKRs. For example, IBMGlobal Services
introducedschemes to identify
and rewardspecific
instances of knowledge contribution (Berry2000).
Organizational reward appears to be particularly

effective under conditions of strong identification.
Thus, to build a critical mass of knowledge
contributors,
can
management
reward and publicize such

employees,
groups,
or
offer organizational
reward firstamong
business
units

2005

whose
133
et al./Contributing
Kankanhalli
to Electronic
Knowledge
Repositories
interests are known to align with those of the

organization.
Fourth,management can raise the perceptions of

reciprocitybenefit among knowledge workers by
highlightingsituationswhere requests forhelp from
knowledge contributors have been promptly
answered. Valued knowledge contributorscan be
asked to testifyinKM events about how theyhave
benefitted from the knowledge contribution of
others. Reciprocity appears to be particularly
important
when
pro-sharing
norms
are
weak.
Alternatively, management can strengthen pro

sharing norms to reduce the necessity of reci
procitybenefit forknowledge contributors toEKR.
Organizations have successfully promoted pro
For
sharing norms through a variety of means.
British
at
Petroleum
office
open
spaces
example,
helped employees to more easily consult each
other (Chiem 2001). General Electric transfers its
employees between departments to promote pro
sharing norms (Dzinkowski 2001).
Finally,management can reduce codificationeffort
by deploying KM systems (including EKRs) that
facilitate entry of knowledge and thereby reduce
the time and effortneeded to codify knowledge.
Commercially
capabilities
available
that prompt
KM
provide
systems
for knowledge
and
auto
matically organize the knowledge captured to

reduce codificationeffort.Examples are taxonomy
and
generators,
classifiers,
more
forms of knowledge
clustering
engines
available fromcompanies such as Autonomy and

Verity. KM systems may be designed to allow
natural
contribution
(e.g.,
or video) as opposed
to purely text
contributions. Besides tryingto reduce the time
and effortneeded to codify knowledge, manage
ment can also allocate time foremployees toshare
knowledge by integratingthis activity intoregular
Such a practice has been
work processes.
audio
common
in consultancy firms like Accenture

al. 1999). Codification effortappears
et
(Hansen
to be a salient barrierwhen generalized trust is
weak. Hence, when it is difficultto deploy KM
manage
systems thatfacilitatecodificationefforts,
ment can raise the levelof generalized trust.This
can be done by giving due credit for knowledge

contributions and ensuring appropriate usage of
134
other's knowledge. Such practices are adhered to

inexemplar KM organizations such as Buckman
Laboratories (Buckman 2004).
The fact that enjoyment in helping others is the

most important
motivator forcontributors to EKRs
followed by knowledge self-efficacy and organi
zational rewards (both directlyand moderated by
identification)impliesa priorityfor themanagerial
recommendations
outlined
above.
For
example,
measures
to increase enjoyment inhelping others

more
be
effective in terms of encouraging
may
EKR contributorsthanorganizational rewards and,
therefore, should be given higher priority.
to increase reciprocity
Similarly, the measures
benefits and reduce codification effortmay be of
lowerpriorityinmotivating contributions to EKRs.
Limitations and Future Studies

Results of this study must be interpreted in the
context of its limitations. First, the use of cross
sectional data and regression analysis do not allow
the possibilityof bidirectional (feedback) effects to
be explored. For instance, the effects of EKR
usage on subsequent perceptions of cost and
benefit factorsby knowledge contributorshas been
recognized
but
cannot
be
examined.
Future
studies can collect longitudinal data to assess

such bidirectional (feedback) effects.
Second, based on a sample of 150 respondents,
several significant results have been obtained.
However, a largersample thatbringsmore statis
ticalpowerwould have allowed more sophisticated

statistical analysis. With such samples, future
studies
can
test a second-order
tural equation modeling

benefit
can
be
modeled
model
techniques.
as
using
struc
Cost and
second-order
con
structswith thevarious cost and benefit factorsas

formative indicators. Such a model would allow a
more rigorous test of the constructs from social
exchange
theory.
Third, our research model was empirically tested

based on the responses of knowledge contributors
from 10 public organizations inSingapore. Since
2005

Kankanhalli
themanner of operation and culture of public and

private (for-profit)
organizations inSingapore isnot
substantially different,the results of our study are
potentially generalizable to private organizations
with KM initiatives. However caution must be
exercised when attempting to generalize the
results across a range of organizations operating
invaried contexts. Future studies can replicate

this study using our revised research model (see
Figure 2) inother contexts. For example, a similar
research approach can be used to investigateEKR
usage by knowledge seekers. Other formsof KM
systems (such as those supporting the personali
zation strategy) can also be studied ina similar
way. Inaddition, our revised research model can
be tested with respondents fromdifferentorgani
zational settings to assess the external validityof
the results.
Fourth,
research
future
can
extend
our
revised
theoreticalmodel (see Figure 2) toaccount forthe

remaining unexplained variance inEKR usage by
knowledge contributors. To extend the revised
theoreticalmodel, additional theoretical perspec
tives such as the technology acceptance model
(Davis 1989) and the task-technology fitmodel
(Goodhue and Thompson 1995) may be incor

porated. The technology acceptance model can
help to better account for the ease of use and
The task-technology fit
usefulness of EKRs.
model can help to better explain the effects of
knowledge type on knowledge contribution to
EKRs. Additionally, the use of concepts fromthe
organizational
memory
information
systems
litera
ture (e.g., Stein and Zwass 1995) may enable

researchers to investigate specific subsystems of
EKRs as well as the KM activities associated with
these
subsystems.
Conclusion
et al./Contributing
the contexts under which
these effects may

level (cost
individual
operate. By encompassing
and benefit) factors and community level (contex
Knowledge
Repositories
tual) factors, this model improves explanatory

power when compared to previous studies. Ina
broader sense, thisstudy demonstrates the value
of using social exchange theoryand social capital
theory to account for the usage of collective
technologies (which includeEKRs as well as other
KM systems).
Besides contributing to theory
building inthe area of KM ingeneral and EKRs in
particular, the results of this study offer useful
implications toKM practitioners.
Ina futurecharacterized by volatile environments,
effective leverage of organizational knowledge
more successful
would be a factor differentiating
from less successful organizations. As a step
toward facilitatingknowledge leverage, knowledge
contributions toEKRs need to be encouraged. As
invest more resources in KM
organizations
initiatives, it is imperative that research on KM
initiativesand KM systems, such as this study,
continue to generate findings that informpractice.
References
Adler, P. S.
"Market,Hierarchy, and Trust: The

Knowledge Economy and the Future of
Capitalism," Organization Science (12:2), 2001,
pp. 215-234.
Agarwal, R. "IndividualAcceptance of Information

Technologies," inFraming the Domains of IT
Management: Projecting theFuture throughthe
Past, R. W. Zmud (Ed.), Pinnaflex Education
Resources,
Aiken,
L. S.,
Cincinnati,
and West,
OH,
S. G.
2000,
Multiple
pp.
85-104.
Regression:
Testing and Interpreting Interactions, Sage

Publications, Newbury Park, CA, 1991.
Alavi, M. "Managing Organizational Knowledge,"
inFraming the Domains of IT Management:
Projecting the Future through the Past, R. W.
Zmud (Ed.), Pinnaflex Education Resources,
Cincinnati, OH, 2000, pp. 15-28.
Alavi,
This study develops and tests a theoreticalmodel

that explains how cost and benefit factors can
impactEKR usage by knowledge contributors,and
to Electronic
M., and
Leidner,
D. E.
"Knowledge
Manage
ment Systems: Issues, Challenges and Bene

fits,"Communications of AIS (1), 1999, pp. 1
37.
Alavi, M., and Leidner, D. E. "Review: Knowledge

Management and Knowledge Management
Systems: Conceptual Foundations and Re

2005

135
Kankanhalli
et al./Contributing
to Electronic
Repositories
Knowledge
search Issues,"MIS Quartedy (25:1), 2001, pp.

107-136.
Ba,
S.,
Stallaert,
J.,
and
Whinston,
A.
B.
"Research Commentary:
Introducinga Third
Dimension in Information
Systems Design?The
Case for IncentiveAlignment," InformationSys
tems Research (12:3), 2001, pp. 225-239.
Babcock, P.
"Shedding Light on Knowledge
Management," HR Magazine
(49:5), 2004, pp.
46-50.
Bandura, A. Social Foundations of Thought and

Action: A Social Cognitive Theory, Prentice
Hall, Englewood Cliffs,NJ, 1986.
Baumeister, R. F. "ASelf-Presentational View of
Social Phenomena," Psychological Bulletin
(91:1), 1982, pp. 3-26.
Beer, M., and Nohria, N. "Cracking the Code of
Change," Harvard Business Review (78:3),
2000, pp. 133-141.
Berry, J. "Employees Cash
inon KM?Knowledge
Management Programs Pay Rewards toShare
Ideas," Internetweek (814), 2000, pp. 45-46.
Blau, P. M. Exchange and Power inSocial Life,
JohnWiley, NewYork, 1964.

Bock, G. W., and Kim, Y. G. "Breaking theMyths
of Rewards: An Exploratory Study ofAttitudes
about
Information
Knowledge
Sharing,"
Resource Management
14-21.
pp.
Bock,
Journal (15:2), 2002,
G. W.,
J. N.
Zmud,
"Behavioral
R. W.,
Kim,
Intention
Y.
G.,
Formation
and
Lee,
inKnowl
edge Sharing: Examining theRoles of Extrinsic

Motivators, Social-Psychological Forces, and
Organizational Climate," MIS Quarterly (29:1),
March 2005, pp. 87-111.
Brown,
J. S.,
and
Duguid,
P.
"Organizational
Learning and Communities of Practice," Organi

zation Science (2:1), 1991, pp. 40-57.
Buckman, R. H. Building a Knowledge-Driven
Organization, McGraw Hill,NewYork, 2004.
"On the Various and Changing
Cheney, G.
Meanings of Organizational Membership: A
Field Study of Organizational Identification,"

Communication Monographs (50), 1983, pp.
342-362.
Chiem, P. X. "Form Follows Strategy,"Knowledge

Management Magazine, October 2001 (avail
able online at http://www.destinationkm.com/
articles/default.asp?ArticlelD=369).
136
Cohen,
and
D.,
L.
Prusak,
In Good
Company:
How Social Capital Makes Organizations Work,

Harvard Business School Press., Boston, 2001
Coleman, J. Foundations of Social Theory,
Harvard University Press, Cambridge, MA,
1990.
Comrey, A. L. A First Course inFactor Analysis,
Academic Press, NewYork, 1973.
Connolly,
T.,
and
B.
Thorn,
K.
"Discretionary
Theory, Data, and Implications," in

Organizations and Communication Technology.
J. Fulk and C. Steinfield (Eds.), Sage Publi
cations, Newbury Park, CA, 1990, pp. 219-233.
Databases:
Constant,
D.,
Kiesler,
S.,
and
Sproull,
L.
"What's
Mine Is Ours or Is It? A Study of Attitudes

about
Information Sharing,"
Information
Systems Research (5:4), 1994, pp. 400-421.
Constant,
D.,
and
L.,
Sproull,
S.
Kiesler,
"The
Kindness of Strangers: The Usefulness of

Electronic Weak Ties for Technical Advice,"
Organization Science (7:2), 1996, pp. 119-135.
Cook,
M.,
and
D. T. Quasi-Experimen
Campbell,
tation: Design and Analysis Issues forField

Settings, Houghton Mifflin,Boston, 1979.
Cronbach, L. J. "CoefficientAlpha and the Internal
Structure of Tests," Psychometrika (16), 1951,
pp. 297-334.
R., and Baird, L. "Technology Is Not

Enough:
ImprovingPerformance by Building
Organizational Memory," Sloan Management
Review (4V.3), 2000, pp. 69-78.
Cross,
T. H., De
Davenport,
Long,
D. W.,
and
Beers,
M.
C. "Successful Knowledge Management Pro

jects,"Sloan Management Review (39:2), 1998,
pp. 43-57.
Davenport,
T. H., and
Prusak,
L. Working
Knowl
edge: How Organizations Manage What They

Know, Harvard Business School Press, Boston,
1998.
Davis, F. D. "Perceived Usefulness, Perceived
Ease of Use, and User Acceptance of Informa
tionTechnology," MIS Quarterly (13:3), 1989,
pp. 319-341.
Deci,
E.
L.,
and
Ryan,
R.
M.
"The
Empirical
Motivational Processes."
Exploration of Intrinsic
inAdvances inExperimental Social Psychology.
L. Berkowitz (Ed.), Academic Press, NewYork,
1980, pp. 39-80.
Dooley, D. Social Research Methods, Prentice
Hall, Upper Saddle River, NJ, 2001.
2005

Kankanhalli
to
"Removing Boundaries
R.
Dzinkowski,
Learning,"Knowledge Management Magazine,

May 2001 (available online at http://www.
destinationkm.com/articles/default.asp7Article
ID=240).
Falk, R. F., and Miller, N. B. A Primer forSoft
Modeling, University of Akron Press, Akron,
OH, 1992.
Fulk,
J., Flanagin,
A.,
M., Monge,
Kalman,
P. R.,
and Ryan, T.
"Connective and Communal
in Interactive Communication
Public Goods
Systems," Communication Theory (6:1), 1996,
pp. 60-87.
P.
Goodman,
and
S.,
E.
Darr,
D.
"Computer
Aided Systems and Communities: Mechanisms

for Organizational Learning in Distributed
Environments,"MIS Quarterly (22:4), 1998, pp.
417-440.
D.
Goodhue,
and
L.,
R.
Thompson,
L.
"Task
Technology Fit and Individual Performance,"

MIS Quarterly (19:2), 1995, pp. 213-236.
Gray, P. H. "The Impact of Knowledge Reposi
tories on Power and Control intheWorkplace,"
Information Technology and People
(14:4),
2001, pp. 368-384.

Green, G. I. "Perceived Importance of Systems
Analyst's Job Skills, Roles and Non-Salary
Incentives,"MIS Quarterly (13:2), 1989, pp.
115-133.
Grover,
V.,
and
T.
Davenport,
H.
"General
spectives on Knowledge Management:

tering a Research
Journal
Agenda,"
Per
Fos
of Manage
21.
H.
"Social
for Knowledge
Exchange
paper
Exchange,"
presented at Managing
and Critiques, Uni
Conversations
Knowledge:
of
Leicester
versity
Management Centre, April
10-11,2001 (available online at http://www.soc.
napier.ac.uk/publication/op/getpublication/
publicationid/321908).
Hansen,
M.
T.,
Nohria,
N.,
and
Tierney,
T.
"What's Your Strategy forManaging Knowl

edge?," Harvard Business Review (77:2), 1999,
pp.
106-116.
Hargadon, A. B. "Firms as Knowledge Brokers:

Lessons inPursuing Continuous Innovation,"
California Management Review (40:3), 1998,
pp. 209-227.
D.
Hickson,
R.
E.,
to Electronic
Knowledge
J., Hinings,
C. R.,
Pennings,
J. M.
and
Repositories
C. A., Schenk,
Lee,
Con
"A Strategic
tingenciesTheory of InterorganizationalPower,"
AdministrativeSciences Quarterly (16:2), 1971,
pp. 216-229.
Igbaria, M.,
Parasraman,
and
S.,
J. J.
Baroudi,
"A
Motivational Model of Microcomputer Usage,"

Journal of Management InformationSystems
(13:1), 1996, pp. 127-143.
S.
Jarvenpaa,
L., and
D.
Staples,
S.
"The
Use
of
Collaborative Electronic Media for Information

Sharing: An Exploratory Study of Determi
nants," Journal of Strategic InformationSys
tems (9:2/3), 2000, pp. 129-154.
Jehn,
K. A.,
Northcraft,
G.
B.,
and
Neale,
M. A.
(1999). "WhyDifferences Make a Difference: A

Field Study of Diversity, Conflict, and Perfor
mance in
Workgroups," AdministrativeScience
Quarterly (44:4), 1999, pp. 741-763.
Johnson,
W.
A. M.,
L., Johnson,
and
F.
Heimberg,
"A Primary and Second Order Component

Identification
Analysis of the Organizational
Educational
and
Questionnaire,"
Psychological
Measurement (59:1), 1999, pp. 159-170.
Kalman, M. E. The Effects of Organizational
Commitment and Expected Outcomes on the
Motivation toShare Discretionary Informationin
a Collaborative Database:
Communication
Dilemmas and Other Serious Games, unpub
lished Ph.D. Thesis, University of Southern
California, Los Angeles, CA, 1999.
Kim,
ment InformationSystems (18:1), 2001, pp. 5

Hall,
et al./Contributing
J.,
and
Mueller,
C.
W.
Factor
Analysis:
Statistical Methods and Practical Issues, Sage

Publications, Newbury Park, CA, 1981.
Kollock, P. "The Economies of Online Coopera
tion: Gifts and Public Goods inCyberspace," in
Communities inCyberspace, M. Smith and P.
Kollock (Eds.), Routledge, NewYork, 1999, pp.
220-239.
Krebs, D. L. "Empathy and Altruism," Journal of

Personality and Social Psychology (32:6), 1975,
pp. 1132-1146.
Leonard-Barton, D. Wellsprings of Knowledge:
Building and Sustaining the Source of Innova
tion,Harvard Business School Press, Boston,
1995.
Liebowitz, J., and Beckman, T.
Knowledge
Organizations: What Every Manager Should
Know, CRC Press, Boca Raton, FL, 1998.

2005

137
Kankanhalli
et al./Contributing
to Electronic
Knowledge
Repositories
Markus, M. L. "Towards a Theory of Knowledge

Reuse: Types of Knowledge Reuse Situations
and Factors in Reuse Success," Journal of
Management Information
Systems (18:1), 2001,
pp. 57-94.
McDermott, R. "Why InformationTechnology

Inspired but Cannot Deliver Knowledge Man
Review
agement," California Management

(41:4), 1999, pp. 103-117.
McKeen,
J. D.,
T.,
Guimaraes,
and Wetherbe,
J.
C.
"The Relationship between User Partici

pation and User Satisfaction: An Investigation
of Four Contingency Factors," MIS Quarterly
(18:4), 1994, pp. 427-451.
McKnight,
L. L., and
D. H., Cummings,
Chervany,
N. L. "InitialTrust Formation inNew Organi

zational Relationships," Academy ofManage
ment Review (23:3), 1998, pp. 473-490.

Microsoft. "Expanded MVP Program Puts Knowl
edge, Real-life Experience toWork forMicrosoft
Customers Worldwide," October 23,
2002
at
online
http://www.microsoft.com/
(available
presspass/features/2002/Oct02/10-23mvp.asp).
Mintzberg, H. The Nature of Managerial Work,
Harper & Row, New York, 1973.
Mishra, A. K. "Organizational Responses toCrisis:
The Centrality of Trust," inTrust inOrganiza
tions:
Frontiers
of Theory
and
Research,.
R. M.
Kramer and T. R. Tyler (Eds.), Sage Publica

tions,Thousand Oaks, CA, 1996, pp. 261-287.
Mistzal, B. A. Trust inModern Societies, Polity
Press, Cambridge, England, 1996.
Molm, L. D. Coercive Power inSocial Exchange,
Cambridge UniversityPress, NewYork, 1997.
Moore,
G. C,
and
Benbasat,
I. "Development
of
an Instrumentto Measure the Perceptions of

Adopting an InformationTechnology Innova
tion," Information Systems
1991, pp. 173-191.
Research
(2:3),
Nahapiet, J., and Ghoshal, S. "Social Capital,

IntellectualCapital, and Organizational Advan
tage,"Academy ofManagement Review (23:2),
1998, pp. 242-266.
Nunally, J.C. Psychometric Theory, McGraw-Hill,
NewYork, 1978.
O'Dell, C, and Grayson, C. J. "IfOnly We Knew
What We Know:
138
Identificationand Transfer of
InternalBest Practices," CaliforniaManagement

Review (40:3), 1998, pp. 154-174.
O'Reilly, C, and Chatman, J. "Organizational
Commitment and Psychological Attachment:
The Effects of Compliance, Identification,and
Internalizationon Prosocial Behavior," Journal
of Applied Psychology (71:3), 1986, pp. 492
499.
Orlikowski,W. J. "Learning fromNotes: Organi

zational Issues inGroupware Implementation,"
InformationSociety (9:3), 1993, pp. 237-251.
Osterloh,
M.,
and
B. S.
Frey,
Knowl
"Motivation,
edge Transfer, and Organizational Forms,"

Organization Science (11:5), 2000, pp. 538-550.
Patchen, M. Participation, Achievement, and
Involvement on the Job, Prentice Hall,
Englewood Cliffs,NJ, 1970.
Pfeffer, J. Managing with Power: Politics and
Influence in Organizations, Harvard Business
School Press, Boston, 1992.
"The Prosperous Community:
Putnam, R. D.
Social Capital and Public Life," The American
(4:13), 1993 (available online at
Prospect
http://www.prospect.org/print/V4/13/putnam
r.html).
Rheingold, H. The Virtual Community: Home

steading on theElectronic Frontier,MIT Press,
Cambridge, MA, 2000.
Ruggles, R. "The State of theNotion: Knowledge
in Practice,"
Management
ment Review
Ruppel,
C.
P.,
California
Manage
(40:3), 1998, pp. 80-89.

and
Harrington,
S.
J.
"Spreading
Knowledge through Intranets: An Analysis of

theOrganizational Culture Leading to Intranet
Adoption and Use," IEEE Transactions on

Professional Communications (44:1), 2001, pp.
37-52.
Schultze, U., and Leidner, D. E. 'StudyingKnowl

in InformationSystems
edge Management
Research: Discourses and Theoretical Assump
tions,"MIS Quarterly (26:3), 2002, pp. 213-242.

Sharma,
S.,
Durand,
R.
M.,
and
Gur
Arie,
O.
Analysis of Moderator
Journal of Marketing Research
"Identification and
Variables,"
(18:3), 1981, pp. 291-300.

Smith,
D.
H.
"Altruism,
Volunteers,
and
Volun
teerism,"Journal of VoluntaryAction Research

(10:1), 1981, pp. 21-36.
2005

Kankanhalli
Starbuck, W. H. "Learning by Knowledge-Inten

sive Firms," Journal of Management Studies
(29:6), 1992, pp. 713-740
Stein,
E. W.,
and
V.
Zwass,
"Actualizing
Organi
zational Memory with InformationSystems,"

InformationSystems Research (6:2), 1995, pp.
85-117.
Stone, E. F. Research Methods inOrganizational

Behavior, Goodyear, Santa Monica, CA, 1978.
F., and
E.
Stone,
J. R.
Hollenbeck,
"Some
Issues
Associated with the Use ofModerated Regres

sion," Organizational Behavior and Human
Performance (34), 1984, pp. 195-213.
Thibaut, J.W., and Kelley, H. H. (1986). The
Social Psychology of Groups. New Brunswick,
Transaction
Books.
Tsai, W., and Ghoshal, S. "Social Capital and

Value Creation," Academy of Management
Journal (41:4), 1998, pp. 464-476.
Vallerand, R. J. "Toward a Hierarchical Model of
Intrinsicand ExtrinsicMotivation,"Advances in
Experimental Social Psychology (29), 1997, pp.
271-360.
Venkatesh,
V.,
Extension
and
F. D.
Davis,
"A Theoretical
of
the Technology Acceptance

Model: Four Longitudinal Field Studies," Man
agement Science (46:2), 2000, pp. 186-204.
M.
Wasko,
M.,
and
S.
Faraj,
"It
Is What
One
Does':
Why People Participate and Help
Others inElectronic Communities of Practice,"
Journal of Strategic Information Systems
(9:2/3), 2000, pp. 155-173.
M. M.,
Wasko,
and
Faraj,
S.
"Why Should
IShare?
Examining Knowledge Contribution inNetworks

of Practice, "MIS Quarterly (29:1), 2005, pp. 35
57.
Weill,
M. H.
P., and Olson,
"An Assessment
Contingency Theory of Management

tion Systems,"
Journal
of Management
of the
Informa
Infor
mation Systems (6:1), 1998, pp. 59-85.

Williamson, O. Markets and Hierarchies: Analysis
and Antitrust Implications, Free Press, New
York, 1975.
Yamagishi,
T.,
and
Cook,
K.
S.
"Generalized
Exchange and Social Dilemmas," Social Psy

chology Quarterly (56:4), 1993, pp. 235-248.
Zack, M. H.
"Managing Codified Knowledge,"
Sloan Management Review (40:4), 1999, pp.
45-58.
et al./Contributing
to Electronic
Knowledge
Repositories
About theAuthors
Atreyi Kankanhalli isan assistant professor inthe
Department of Information
Systems at theNational
of
University Singapore (NUS). She obtained her
B. Tech. from the Indian Instituteof Technology
Delhi, M.S. from the Rensselaer
Polytechnic
Institute,New York, and Ph.D. fromNUS. She
has been a Visiting Scholar at the University of
California Berkeley. Professor Kankanhalli had
considerable work experience in industrialR81D

and consulted forseveral organizations including
World Bank. Her work has appeared intheJournal
of theAmerican Society for InformationScience
and Technology, Communications of the ACM,

Decision Support Systems, InternationalJournal of
Information
Management, and the proceedings of
the International Conference on Information
among
Systems
others.
She
serves
on
several
informationsystems conference committees and

on the editorial board of the InternationalJournal
ofKnowledge Management. Professor Kankanhalli
has been awarded the InfocommDevelopment
Authority Gold Medal at NUS and the ACM

SIGMIS
ICIS 2003 Best Doctoral Dissertation
award. Her research interests are in knowledge
virtual
management,
ment,
and
information
teams,
systems
electronic
govern
security.
Bernard C. Y. Tan isan associate professor and

Head of theDepartment of Information
Systems at
the National University of Singapore (NUS). He
received his Ph.D. in InformationSystems from
NUS.
Professor
Tan
has
won
several
research
and teaching awards at NUS. He has been a

Visiting Scholar in the Graduate School of Busi
ness at Stanford Universityand theTerryCollege
of Business at the UniversityofGeorgia. He isan
elected Asia-Pacific Council Member for the
Association for InformationSystems. Professor
Tan serves as a senior editor forMIS Quarterly
and e-Service Journal, and is on the editorial
boards of IEEE Transactions on Engineering
Management, Journal of theAIS, Informationand
Man
Management, Journal of Global Information
agement, and International Journal of Distance
Education Technologies. His research has been
published inACM Transactions on Computer
Human Interaction,ACM Transactions on Infor

2005

139
Kankanhalli
et al./Contributing
to Electronic
Knowledge
Repositories
mation Systems, Communications of the ACM,

Decision Support Systems, European Journal of
Information Systems,
IEEE Transactions on
Engineering Management, IEEE Transactions on
Professional Communication, IEEE Transactions
on Systems, Man, and Cybernetics, Information
and Management, Information
Systems Research,
InternationalJournal ofHuman-Computer Studies,
Journal of Management
Information Systems,
Journal of the AIS, Journal of the American
Society for InformationScience and Technology,

Management Science and MIS Quarterly. His
current
research
focuses
computer-mediated
management,
and
on cross-cultural
issues,
communication, knowledge
information
privacy.
Kwok-Kee Wei isChair Professor and Head of the

Department of Information Systems at City
Universityof Hong Kong. He received his D.Phil.
140
inComputer Science
from the University of York

He is on leave from the
(United Kingdom).
National Universityof Singapore, where he ispro
fessor in the Department of InformationSystems.
He is past President of theAssociation for Infor
mation Systems. He has served on the editorial
boards ofMIS Quarterly and InformationSystems

Research. His research has been published in
ACM Transactions on Computer-Human Interac
tion,ACM Transactions on InformationSystems,
European Journal of InformationSystems, IEEE

Transactions on Systems, Man, and Cybernetics,
Information Systems Research,
International
Journal of Human-Computer Studies, Journal of
Management InformationSystems, Journal of the
AIS, MIS Quarterly, and Management Science.
His
research
focuses
on computer-mediated
com
munication, innovation and knowledge manage

ment,
and
human-computer
2005

interaction.
Kankanhalli
et al./Contributing
to Electronic
Knowledge
Repositories
Appendix
^
Survey Items
Construct
(LOKP)
ItemWording
and Code Source
Sharing my knowledge throughEKRs makes me

lose my unique value in the organization (LOKP1)
lose my power base in the organization (LOKP2)
lose my knowledge thatmakes me stand out with
Orlikowski
respect to others (LOKP3)
lose my knowledge that no one else has (LOKP4)
Developed based on
Thibaut and Kelley
(1986)
Developed based on
Orlikowski (1993)
Developed based on
Ido not have the time to enter my knowledge into
(1993)
Developed based on
Orlikowski (1993)
Developed based on
It is laborious to codifymy knowledge intoEKRs
Developed
based on
The effort is high forme to codifymy knowledge
Developed
based on
EKRs(CEFF1) Orlikowski
(1993)
(CEFF2) Orlikowski
(1993)
Effort
?_._.
int0
EKRs (CEFF3)
Iam worried that ifIshare my knowledge through

*
EKRs, Iwill have to spend additional time
answering followup questions (CEFF4)
Iam afraid thatmy submission to EKRs will evoke
additional clarifications or requests forassistance
(CEFF5)
Organizational
Reward
(OREW)
Orlikowski
(1993)
Developed based on
Goodman and Darr
(1998)
Developed based on
Goodman and Darr
(1998)
It is importantto get a betterwork assignment when

Adapted fromKalman
Ishare my knowledge throughEKRs (OREW1)
(1999)
It is importantto be promoted when Ishare my
Developed based on
knowledge throughEKRs (OREW2)
Hargadon
(1998)
It is importantto get a higher salary when Ishare
Developed based on
Hall
my knowledge throughEKRs (OREW3)
(2001)
It is importantto get a higher bonus when Ishare
Developed based on
Hall
(2001)
It is importantto get more job securitywhen Ishare
Developed based on
Davenport and Prusak
_|_j
(1998)_

2005

141
Kankanhalli
et al./Contributing
to Electronic
Construct
Knowledge
ItemWording
Imaoe
(\td\AC\
'
(
Repositories
and Code Source
Sharing my knowledge throughEKRs improvesmy

imagewithin the organization (IMAG1)
People in the organization who share theirknowledge throughEKRs have more prestige than those
who do not (IMAG2)
#
Sharing my knowledge throughEKRs improves
others recognitionofme (IMAG3) (1989)
When Ishare my knowledge throughEKRs, the
people Iwork with respect me (IMAG4)
When Ishare my knowledge throughEKRs, my
superiors praise me (IMAG5) (1999)
When Ishare my knowledge throughEKRs, I
believe that Iwill get an answer forgiving an answer
(RECP1)
Ishare my knowledge throughEKRs,
When
_
.A
(RECpT
expect
somebody
<RECP2>
<1"3>
to respond
when
(2000)
I
I'm in need
When Icontribute knowledge to EKRs, Iexpect to

get back knowledge when Ineed it(RECP3)
When Ishare my knowledge throughEKRs, I
believe thatmy queries forknowledge will be
answered infuture (RECP4)
Adapted fromMoore
and Benbasat (1991)
Adapted fromMoore
and Benbasat (1991)
Adapted fromGreen
Adapted fromKalman
(1999)
Adapted fromKalman
Developed based on
Wasko and Faraj
Developed
based on
Yamagishi
and
Cook
Developed based on
feedback fromsorters
Developed based on
feedback fromsorters
Ihave confidence inmy ability to provide knowledge

thatothers inmy organization consider valuable
-I
f"Fff
(KSEF1)
have the expertise needed to provide valuable

knowledge formy organization (KSEF2)
Itdoesn't reallymake any differencewhether Iadd
to the knowledge others are likelyto share through
Adapted fromKalman
(1999)
EKRs (KSEF3)
Most other employees can provide more valuable

knowledge than Ican (KSEF4)
Ienjoy sharing my knowledge with others through
EKRs(EHLPI)
Enjoyment
ii i .
in
Helping Others
p^ '
^fhi
(
Ienjoy
a
* my knowledge
s ' helping others by sharing
.,
through
# u.
_/,_
EKRs
,_,,. _0/ ...

(EHLP2)
.. V
,
, u u
p someone e,se by shann9 my
EKRs (EHLP3) (
through
knowledge
_
Developed
Sharing my knowledge with others throughEKRs

gives me pleasure (EHLP4)_
142
based
Wasko and Faraj

(200m}
2005

on
Kankanhalli
to Electronic
Knowledge
Repositories
ItemWording and Code Source
Construct
,rTRI '
^
et al./Contributing
Ibelieve thatpeople inmy organization give credit

forother's knowledge where it isdue (GTRU1)
Ibelieve thatpeople inmy organization do not use
unauthorized knowledge (GTRU2)
Ibelieve thatpeople inmy organization use other's
Developed based on
Mishra (1996)
knowledge appropriately (GTRU3)

Ibelieve thatpeople inmy organization share the
best knowledge that they have (GTRU4)
There isa norm of cooperation inmy organization
Developed based on
Goodman
and Darr
(PSNM1)
There isa norm of collaboration inmy organization
Pro-Sharin
(PSNM)
Identification
(IDEN)
(PSNM2)
(1998)
There isa norm of teamwork inmy organization
(PSNM3)
(1998)
Developed based on
Goodman and Darr
Developed
based on
Starbuck1992
There isa willingness to value and respond to

Developed based on
Leonard-Barton 1995
diversity inmy organization (PSNM4)
There is a norm of openness to conflictingviews in
Developed based on
Leonard-Barton
1995
my organization (PSNM5)
There isa norm of tolerance ofmistakes inmy
Developed based on
Leonard-Barton
1995
organization (PSNM6)
Iam glad Ichose towork forthisorganization rather
than another company (IDEN1)
I talkof thisorganization tomy friendsas a great
company towork for (IDEN2)
Iam willing to put ina great deal of effortbeyond
thatnormallyexpected to help my organization to
be successful (IDEN3)
I find thatmy values and my organization's values
are very similar (IDEN4)
Adapted
Ingeneral the people employed bymy organization
are working toward the same goal (IDEN5)
I find iteasy to identify
myself with my organization
1983
fromCheney
(IDEN6)
I feel thatmy organization cares about me (IDEN7)

I feel a sense of belonging towards my organization
(IDEN8)
Iam proud to be an employee of thisorganization
(IDEN9)
EKR Usage
(EUSG)
What isyour frequency of usage of.EKRs to

contribute knowledge? (EUSG1)
al.
Ioften use EKRs to contributemy knowledge inmy
work.(EUSG2) (1989)
I regularlyuse EKRs to contributemy knowledge in

work. (EUSG3) (1989)
my
_
Adapted from Igbaria et

(1996)
Adapted fromDavis
Adapted fromDavis

2005

143

PDF

Uploaded by

Document Information

Original Description:

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

PDF

Uploaded by

Copyright:

Available Formats

Contributing Knowledge to Electronic Knowledge Repositories: An Empirical Investigation

Author(s): Atreyi Kankanhalli, Bernard C. Y. Tan and Kwok-Kee Wei

This content downloaded from 130.102.158.12 on Fri, 14 Nov 2014 15:15:49 PM

By: Atreyi Kankanhalli

Organizations are attempting to leverage their

large number of KM initiatives fail due to the

for the moderating

The model is validated through a large-scale

by knowledge contributors. Contextual factors

cation) moderate the impact of codification effort,

textual factors whereas the effects of intrinsic

by contextual factors. The loss of knowledge

This content downloaded from 130.102.158.12 on Fri, 14 Nov 2014 15:15:49 PM

usage by knowledge contributors. Besides contri

Knowledge management isdefined as "a systemic

have been identified in the informationsystems

facilitate knowledge reuse throughaccess to the

technological capabilities are important,

1998; O'Dell and Grayson 1998).

This study focuses on EKRs since they are

expertise) and people seeking knowledge from

This content downloaded from 130.102.158.12 on Fri, 14 Nov 2014 15:15:49 PM

EKRs for reuse. Success of EKRs requires that

the firststep toward knowledge leverage through

1993; Wasko and Faraj 2000) in attempts to

Constant et al. 1996; Jarvenpaa and Staples 2000;

Existing empirical studies focus on the benefits

suggestions frompractitioner literature(O'Dell and

knowledge contribution in two importantways.

edge contributors. In investigating the usage of

Although the technology acceptance model may

by knowledge contributors that may affect their

and Darr 1998; Jarvenpaa and Staples 2000;

Cost and Benefit Factors

exchanges (Blau 1964), which differ from eco

existence of relatively long-term relationships of

(Fulk et al 1996) where more than two people

Theoretical Background 1HHH

model has been suggested

MIS QuarterlyVol. 29 No. 1/March

This content downloaded from 130.102.158.12 on Fri, 14 Nov 2014 15:15:49 PM

with the EKR serving as the intermediarybetween

expectation of future return. The quantity and

assumption of relatively longer-termrelationships

(tangible and intangible) are the cur

during social exchange or negative outcomes of

received as a resultof social exchange or positive

Social exchange theoryposits thatpeople behave

knowledge sharing using KM systems (Markus

During social exchange, costs can be incurred in

chosen. For example, the timeand effortrequired

rewards for theircontributions (Beer and Nohria

Constant et al. 1994; Constant et al. 1996), which

sharing theirknowledge, contributorsmay receive

edge self-efficacyor confidence in theirability to

help others (Ba et al. 2001; Wasko and Faraj

acts of helping others (Baumeister 1982; Krebs

(Vallerand 1997), including

Contextual factors inour study are derived from

capital theoryposits thatsocial capital provides the

define the context for knowledge exchange are

social relationships thatcan improve the efficiency