ArticlePDF Available

A Multilevel Investigation of the Motivational Mechanisms Underlying Knowledge Sharing & Performance

Authors:

Abstract and Figures

This study draws on three different, yet complementary, theories of motivation, which we combine in an interactive manner, to explain the mechanisms that underlie the exchange between knowledge providers and recipients and ultimately impact performance. More specifically, we use incentive, goal-setting-social cognitive, and social motivation theories to examine knowledge sharing within dyads and its influence on individual performance. One hundred and twenty participants functioning as interdependent manager dyads completed a strategic decision-making simulation. Hierarchical regression and random coefficient modeling techniques were used to test hypothesized relationships. Results demonstrated that the effect of group-oriented incentive systems on the knowledge provider was enhanced when more positive norms for knowledge sharing existed among dyad members. The recipient's self-efficacy had a stronger relationship with performance goals when the recipient trusted the provider. Finally, self-set goals and knowledge sharing had both direct and interactive effects on individual performance. We argue that these findings constitute a useful advance in middle-range motivation (Landy and Becker 1987, Pinder 1984) theory pertaining to knowledge sharing and utilization.
Content may be subject to copyright.
Organization Science
Vol. 18, No. 1, January–February 2007, pp. 71–88
issn 1047-7039 eissn 1526-5455 07 1801 0071
informs®
doi 10.1287/orsc.1060.0223
© 2007 INFORMS
A Multilevel Investigation of the Motivational Mechanisms
Underlying Knowledge Sharing and Performance
Narda R. Quigley
Villanova School of Business, Villanova University, 800 Lancaster Avenue, Villanova, Pennsylvania 19085,
narda.quigley@villanova.edu
Paul E. Tesluk, Edwin A. Locke, Kathryn M. Bartol
Robert H. Smith School of Business, University of Maryland, College Park, Maryland 20742
{ptesluk@rhsmith.umd.edu, elocke@rhsmith.umd.edu, kbartol@rhsmith.umd.edu}
This study draws on three different, yet complementary, theories of motivation, which we combine in an interactive man-
ner, to explain the mechanisms that underlie the exchange between knowledge providers and recipients and ultimately
impact performance. More specifically, we use incentive, goal-setting-social cognitive, and social motivation theories to
examine knowledge sharing within dyads and its influence on individual performance. One hundred and twenty participants
functioning as interdependent manager dyads completed a strategic decision-making simulation. Hierarchical regression and
random coefficient modeling techniques were used to test hypothesized relationships. Results demonstrated that the effect
of group-oriented incentive systems on the knowledge provider was enhanced when more positive norms for knowledge
sharing existed among dyad members. The recipient’s self-efficacy had a stronger relationship with performance goals
when the recipient trusted the provider. Finally, self-set goals and knowledge sharing had both direct and interactive effects
on individual performance. We argue that these findings constitute a useful advance in middle-range motivation (Landy
and Becker 1987, Pinder 1984) theory pertaining to knowledge sharing and utilization.
Key words: knowledge sharing; motivation; performance; multilevel model
Scholars and practitioners alike have increasingly re-
garded an organization’s ability to facilitate the sharing
and utilization of knowledge as critical for organiza-
tional effectiveness (cf. Bock and Kim 2002, Kogut and
Zander 1996, Nonaka and Takeuchi 1995, Tsai 2001).
Particularly in the emerging structure of distributed orga-
nizations, effectiveness is dependent on how well knowl-
edge is shared between individuals, teams, and/or units
(Alavi and Leidner 2001, Argote et al. 2000, Good-
man and Darr 1998, Pentland 1995). Growing evidence
suggests that organizations are more productive when
they are able to successfully create the conditions in
which knowledge is shared by potential providers and
then actively put to use by the recipients of new knowl-
edge (Argote et al. 1990, Baum and Ingram 1998). It
is not surprising, therefore, that scholars from several
different fields have actively studied knowledge manage-
ment and its relationships to effectiveness. For exam-
ple, strategy and organizational learning researchers have
considered the knowledge management problem at the
firm level (e.g., Cole 1998, Kogut and Zander 1996,
Nonaka and Takeuchi 1995, Spender and Grant 1996),
while information technology (IT) and organizational
design researchers have considered organizations as sys-
tems that enable knowledge exchange (e.g., Alavi and
Leidner 2001, Pentland 1995). For their part, organi-
zational behavior researchers and social psychologists
have considered interpersonal issues and the role of
group dynamics in knowledge management (e.g., Bartol
and Srivastava 2002, Moreland and Myaskovsky 2000,
Stasser and Titus 2003).
In their attempts to develop a clearer understanding of
the knowledge sharing and utilization process between
providers and recipients, researchers across these differ-
ent disciplines have noted the critical role of motivation
(e.g., Argote and Ingram 2000; Alavi and Leidner 2001;
Hansen 1999; Goodman and Darr 1998; Spender and
Grant 1996; Szulanski 1996, 2000). For instance, prior
work has suggested that the motivation of knowledge
providers is important for engaging in the effort and
time required to transfer knowledge and overcome con-
cerns about ownership of information (Davenport and
Prusak 1998, Goodman and Darr 1998, Hansen et al.
2005, Kostova 1999). Similarly, researchers have begun
to examine how motivational factors influence the extent
to which recipients seek out, accept, and utilize exter-
nal knowledge (Hayes and Clark 1985, Katz and Allen
1982, Levin and Cross 2004, Mayer et al. 1995, Srinivas
2000, Szulanski 1996).
However, despite the emphasis on and interest in moti-
vational factors when studying the knowledge sharing
and utilization process, there are several limitations
of the existing literature. First, there are no coher-
ent, integrated, theoretical frameworks of the motiva-
tional factors that explain how knowledge is transferred
between knowledge providers and recipients and then is
71
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
72 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
utilized in ways that benefit performance. For instance,
researchers interested in predicting knowledge sharing
have used concepts from social motivation theory, such
as trust, to help explain knowledge transfer (e.g., Levin
and Cross 2004), while others have relied more on
reward and incentive theory to study the impact of incen-
tives on knowledge sharing (e.g., Kalman et al. 2002);
yet no systematic attempts have been made to either
compare or integrate these different potential motiva-
tional mechanisms that explain knowledge sharing. Con-
sequently, we do not know whether knowledge sharing is
better predicted when different motivational perspectives
are included together, as would be the case in study-
ing how rewards and social motivation factors, such as
norms for sharing, may combine to predict knowledge
transfer. Furthermore, no studies have looked at the rele-
vance of goal setting and social cognitive theories in pro-
moting knowledge utilization, despite research showing
that utilizing relevant knowledge requires setting high-
performance goals (Chesney and Locke 1991).
A second limitation of the existing literature is that we
know very little about how motivational factors identified
with knowledge providers and recipients work in con-
junction with each other, because the motivational mech-
anisms across these domains are rarely studied together
(Levin and Cross 2004, Szulanski 1996). Consequently,
questions regarding how motivational variables interact
and the different roles they play in affecting knowledge
sharing and performance have not been systematically
studied. Variables from different theoretical perspectives
will not necessarily work together additively and differ-
ent variables may have effects in different parts of any
integrated model. For example, incentives might affect
knowledge sharing, whereas goals may affect knowledge
utilization. Pinder (1984) and Landy and Becker (1987)
refer to the process of considering the combined and
interactive impact of multiple theories as new “middle-
range” motivational theory development. In the case of
Figure 1 Overview of Framework with Hypothesized Relationships
Incentive condition
Self-efficacy
Knowledge
shared by
knowledge
provider
Self-set goal
Performance
Norms
Knowledge × goal
Trust
H1
H2
H3
Perspective of knowledge provider
Perspective of knowledge recipient
Note. Shading indicates variables at dyad level.
knowledge sharing, a new middle-range theory would
focus on predicting a specific set of outcomes such as
knowledge transfer and utilization by including multiple
theoretical perspectives.
A third limitation of the existing research is that al-
though motivational mechanisms involved in knowledge
transfer and utilization have been identified as stemming
from factors at the individual, social, and organizational
levels of analysis (Szulanski 1996, 2000), most research
has focused on one level of analysis at a time. Much of it
has been done at the organizational level (e.g., Goodman
and Darr 1998), making it difficult to directly examine
specific motivational mechanisms that influence actual
knowledge exchange and use (Argote and Ingram 2000).
Single-level approaches potentially overlook important
multilevel relationships, such as the influence that orga-
nizational factors (e.g., incentive practices) may have on
the individual-level motivation of potential knowledge
providers or the influence that emergent properties at the
group level (e.g., norms) may have on the individual-
level behavior of potential knowledge providers.
In summary, prior research exploring the motivational
factors that help explain the transfer and utilization
of knowledge between potential providers and recipi-
ents has lacked an integrated theoretical framework, has
rarely considered both providers and recipients simulta-
neously as they are engaged in the process of knowledge
exchange and utilization, and has generally neglected
multilevel impacts. We begin to address these issues in
this study by drawing upon three complementary moti-
vational theories and examining how they interact to
explain the motivational mechanisms underlying knowl-
edge transfer and utilization. In doing so, we incorporate
variables that reside at multiple levels of analysis.
Theoretical Model
The theoretical model we develop (shown in Figure 1)
reflects a multilevel view of the process of knowledge
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
Organization Science 18(1), pp. 71–88, © 2007 INFORMS 73
exchange between providers and recipients and is de-
rived from a review of the motivation and knowledge-
sharing literatures. Our review suggests that there are
at least three motivational perspectives that are useful
when considering knowledge sharing by providers and
use of shared knowledge by recipients. Further, we argue
that there are several places where these theories can
be combined in an interactive form to better explain
knowledge transfer and utilization. This approach has
the potential to more effectively account for variance in
motivational outcomes than if the theories were simply
combined in an additive fashion.
Researchers have noted that a critical consideration
is whether the source of new knowledge (the provider)
is motivated to cooperate with the potential recipient
(Szulanski 1996, 2000). Incentive theory (Lawler 1981)
is a useful motivational perspective to draw upon to
study the potential knowledge provider’s decision to
share knowledge because it has been found to be most
predictive when studying decisions involve choosing be-
tween alternative actions (e.g., to share knowledge or
not) and determining the amount of effort allocated to
those behaviors (Landy and Becker 1987, Pinder 1984).
In our theoretical model, we use incentive theory to
investigate whether team-based—as compared to indi-
vidually based—rewards influence the extent to which
knowledge providers share their knowledge with recipi-
ents.
However, researchers have suggested that incentives
provided to encourage knowledge exchange fail to
explain knowledge-sharing behavior adequately (e.g.,
Brown and Duguid 1991, 2002). Instead, the social con-
text that arises through interactions between knowledge
providers and recipients (e.g., in the form of shared
expectations for sharing and exchanging potentially use-
ful knowledge) may be necessary to reinforce incentives.
Social motivation theory (Geen 1991), derived from
early work on peer influences (e.g., Roethlisberger and
Dickson 1939), suggests that motivational mechanisms
such as norms that result from interpersonal interactions
likely have an indirect influence on individual motiva-
tion by intensifying other existing motivational effects
(Geen 1991). Though prior research has suggested that
norms for knowledge sharing are important in encourag-
ing providers to share their knowledge (e.g., Davenport
and Prusak 1998, Jarvenpaa and Leidner 1999), norms
have not been studied in terms of how they may interact
with incentives to strengthen the motivational effect of
team-based rewards on knowledge-sharing behavior.
Beyond knowledge sharing, being able to predict per-
formance requires understanding the motivational mech-
anisms that underlie knowledge recipients’ willingness
to apply the new knowledge they have acquired in ways
that promote performance (Szulanski 1996, 2000). Goal-
setting and social cognitive theory (Bandura 1997, Locke
and Latham 1990) are particularly useful in predicting
performance and understanding the cognitive mecha-
nisms involved in setting performance goals (Landy and
Becker 1987, Miner 2004, Pinder 1984). While the rela-
tionship between self-efficacy (i.e., task-specific self-
confidence, Bandura 1997) and self-set goals, and the
subsequent influence self-set goals have on performance,
have been well established in prior literature (Bandura
1997, Locke and Latham 1990), researchers have sug-
gested that other factors besides the recipient’s goals
influence the effectiveness of knowledge transfer (Szu-
lanski 2000). Specifically, the degree to which recipients
trust their knowledge providers is important both in the
way trust decreases the perceived cost of exchanging
knowledge (Zaheer et al. 1998) and increases the will-
ingness of knowledge providers to apply new knowledge
(Dirks and Ferrin 2001). Although the extent to which
recipients trust their knowledge providers has been
found to be important in predicting transfer (Szulanski
et al. 2004), no attempts have been made to understand
the motivational role trust may play in influencing the
degree to which recipients’ self-efficacy translates into
performance goals. Drawing from social motivation the-
ory and examining its interaction with goal-setting/social
cognitive theory, we address this question.
While the knowledge transfer literature has stressed the
importance of recipients being motivated to apply avail-
able new knowledge (Szulanski 1996, 2000), research
has yet to examine how recipients’ performance objec-
tives work in conjunction with access to new knowl-
edge from providers to impact performance. An extensive
literature in goal-setting theory has shown that both
motivation and subsequent performance increase when
individuals set higher goals for themselves (Locke and
Latham 1990). Although previous studies have shown
that goal effects depend on knowledge (Locke and
Latham 2002), no previous study has looked at the inter-
action between goals and knowledge sharing by peers.
In the final component of our model, we develop and
test theory for the interaction between recipient goals
and knowledge sharing by providers in predicting perfor-
mance. The model summarized in Figure 1 is an integra-
tion of three motivational perspectives—incentive theory,
goal-setting/social cognitive theory, and social motiva-
tional theory—each selected because of its applicability
in understanding a specific aspect of the knowledge-
sharing and transfer process between providers and recip-
ients. Next, we develop the specific hypotheses that form
the basis for the model we test.
Hypothesis Development
The Interactive Influence of Incentives and
Norms on Shared Knowledge
A critical challenge in the knowledge transfer process is
in motivating potential sources to share their knowledge
with recipients. Research suggests that both incentives to
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
74 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
cooperate with recipients and norms that support knowl-
edge sharing are important in motivating providers to
engage in the effort required to transfer their knowl-
edge to others. For instance, researchers have found that
team-oriented incentive structures can facilitate cooper-
ation, thereby enabling organizations to use incentives
to help resolve the classic social dilemma associated
with knowledge sharing that knowledge senders face
(e.g., Dawes 1980, Ferrin and Dirks 2003, Hackman
1987, Kalman et al. 2002, Lawler 1981, Messick and
Brewer 1983, Rutte and Wilke 1992, Van Lange et al.
1992). Group-based incentive structures highlight group
member interdependencies, thus motivating potential
knowledge providers to engage in cooperative behav-
iors such as sharing knowledge in pursuit of higher
group performance and greater rewards (Hackman 1987,
Lawler 1981, Wageman 2001, Wageman and Baker
1997). In this sense, incentive structures can be viewed
along a continuum of outcome interdependence, rang-
ing from purely group-based to purely individual-based
systems (Wageman 2001), with incentives that combine
group and individual components (“hybrid” systems;
e.g., Bartol and Durham 2000, Kandel and Lazear 1992,
Wageman 1995) in between. Consistent with this con-
tinuum, and in keeping with incentive theory, we expect
knowledge providers working under hybrid systems to
share more knowledge than those working under purely
individual-based incentives, though they will not share
as much knowledge as those working under pure group-
based incentives.
Although incentives may play a role in motivating
providers to share their expertise, knowledge exchange is
a social process (Alavi and Leidner 2001, Bock and Kim
2002, Holtshouse 1998, Levin and Cross 2004). Thus,
individuals engaging in knowledge sharing are likely to
be influenced by social motivation factors that emerge
through the exchanges and interactions that potential
providers and recipients have with each other (e.g., Geen
1991, Hackman 1992). Brown and Duguid’s (1991,
1998, 2001, 2002) work has highlighted how socially
developed norms within “communities of practice” are
critical and affect organizational knowledge sharing.
Norms are defined as shared behavioral expectations that
develop over repeated interactions between individuals
and become relatively stable over time (Bettenhausen
and Murningham 1985). Existing research and theory
suggest that norms regarding knowledge sharing that
develop within teams may influence team communica-
tion and knowledge sharing (e.g., Argote et al. 2003,
Barry and Crant 2000, Goodman and Darr 1996). Strong
norms can help knowledge senders overcome the per-
ceived costs incurred in taking time and expending effort
to share information (Borgatti and Cross 2003).
While in keeping with prior literature, we expect
both group-based incentives and norms to have direct
influences on knowledge sharing as described above, a
more novel and, as of yet, untested prediction derived
from social motivation theory (Geen 1991) is that
norms and incentives interact to serve as a strong
influence on knowledge-sharing behavior. As articulated
by Geen (1991), social motivation theory argues that
social influence mechanisms such as norms can influ-
ence behaviors by serving to intensify or strengthen
the motivational tendencies of structural features such
as incentives. Specifically, norms that support knowl-
edge sharing may accentuate the influence that team-
based incentives have on knowledge-sharing behavior by
making cooperation highly salient to providers. While
the notion of aligning norms so that they reinforce
team-based incentives has been touched on within the
practitioner literature (e.g., Davenport and Prusak 1998,
Goodman and Darr 1996), it has not been previously
articulated theoretically.
Prior research has shown that group norms can have
an amplification effect by highlighting certain social and
structural group conditions (Hackman 1992). This was
demonstrated in the work of Deutsch and his colleagues
(Deutsch 1949; Deutsch et al. 1967, 1969), who inves-
tigated the effects of cooperative and competitive incen-
tive structures on cooperation. Hornstein and Deutsch
(1967, p. 312) observed that even when incentives are in
place to promote cooperation, team members only coop-
erate when they perceive “that one’s choice to cooperate
will be reciprocated, permitting cooperation to be real-
ized.” More recent findings that group-based incentives
often fail to yield greater cooperation and higher levels
of performance (Wageman 1995, 2001; Wageman and
Baker 1997) are consistent with these earlier observa-
tions and suggest that norms supporting knowledge shar-
ing may be an important potential moderator of the influ-
ence of group-based incentives on the extent to which
knowledge senders share knowledge. Norms support-
ing knowledge sharing may help overcome the oppor-
tunity costs and risks involved when potential knowl-
edge senders consider communicating new and poten-
tially useful knowledge to colleagues (Bartol and Srivas-
tava 2002, Jarvenpaa et al. 1998).
It is important to note that although there is a strong
theoretical basis to consider the interactive effects of
incentives and social motivation on knowledge sharing,
no prior empirical work has directly addressed this asser-
tion by explicitly integrating both theories and testing
this potential interaction. Indeed, we expect that the con-
sideration of incentives and social motivation in inter-
active combination will have an important impact on
knowledge sharing, demonstrating the limitations of con-
sidering merely the additive direct effects of incentives,
and norms on knowledge sharing. Therefore, we expect:
Hypothesis 1 (H1). The norms supporting knowl-
edge sharing that develop between a knowledge sender
and recipient will moderate the relationship between
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
Organization Science 18(1), pp. 71–88, © 2007 INFORMS 75
incentives and knowledge shared by the sender such that
the positive relationship between incentives and knowl-
edge shared will be stronger when norms supporting
knowledge sharing are more positive.
The Interactive Influence of Self-Efficacy and
Trust on Recipient Choice of Goal Level
The question of how organizations can encourage the
application of available knowledge remains underexam-
ined (Alavi and Leidner 2001). While there are many
reasons why individuals do not apply available knowl-
edge (e.g., causal ambiguity and absorptive capacity,
Cohen and Levinthal 1990), researchers have pointed
to the importance of motivational factors (Hayes and
Clark 1985, Katz and Allen 1982, Levin and Cross
2004, Mayer et al. 1995, Srinivas 2000). In particular,
the performance goals of the recipient have been identi-
fied as important in influencing recipients’ motivation to
use available new knowledge (Gupta and Govindarajan
2000, Szulanski 1996). So far, however, the literature
has not considered the specific mechanisms by which
recipients’ performance goals influence this motivation.
We assert that social cognitive and goal-setting theories
(Bandura 1997, Locke and Latham 1990) offer useful
insights into these processes.
Both social cognitive and goal-setting theories have
suggested that individuals with higher levels of self-
efficacy are more likely to perform well and to set
higher performance goals (cf. Bandura 1997, Locke and
Latham 1990). Self-efficacy refers to individuals’ judg-
ments of their own capabilities to organize and exe-
cute courses of action to successfully perform a given
task (i.e., task-specific self-confidence, Bandura 1997).
Although self-efficacy has been applied to groups, it
has not been applied to situations in which knowledge
sharing between group members is required. In complex
decision-making situations in which individuals must
rely on exchanging information with others to access
knowledge for successful task accomplishment, how-
ever, the degree to which these interpersonal working
relationships are based on trust may influence how self-
efficacy translates into goal setting.
Trust is often defined as a belief that another individ-
ual makes efforts to fulfill commitments, is honest, and
does not seek to take unfair advantage of opportunities
(e.g., Cummings and Bromiley 1996, Dirks and Ferrin
2001, Zucker 1987). Trust has been found to facilitate
knowledge sharing in a variety of settings involving team
member interdependence (e.g., Jarvenpaa and Leidner
1999, Jarvenpaa et al. 1998, Moreland and Myaskovsky
2000, Rau 2005). Given the ambiguity inherent and
effort required in the knowledge transfer process (e.g.,
Szulanski 1996, Szulanski et al. 2004), it is not surpris-
ing that recent research has shown trust to be impor-
tant in facilitating knowledge utilization (e.g., Dirks and
Ferrin 2001, Szulanski et al. 2004, Zaheer et al. 1998).
Although the extent to which recipients trust their
knowledge providers has been found to be important
in predicting transfer, no attempts have been made to
understand the motivational role trust may play in in-
fluencing the degree to which recipients’ self-efficacy
translates into performance goals. Trust typically car-
ries an expectation that one can rely on another person’s
actions and words and that the other person has good
intentions toward oneself (e.g., Dirks and Ferrin 2001,
Mayer et al. 1995, McAllister 1995, Robinson 1996). As
an expectation, trust affects how one assesses the future
behavior of another party with whom one is interde-
pendent; this assessment may have a critical impact on
motivation.
No prior research has empirically tested this idea.
Dirks (1999) and Dirks and Ferrin (2001) argued that,
theoretically, trust may moderate many of the moti-
vational mechanisms that influence performance. They
posited that when an individual’s performance requires
cooperation from others, a lack of trust in being able
to obtain that assistance is likely to decrease motiva-
tion. In addition, trust is likely to affect the extent to
which the individual believes in the veracity of the
shared knowledge, and is therefore willing to act on it
(McEvily et al. 2003, Szulanski et al. 2004). Trust may
have a similar influence on what knowledge recipients
believe about the utility and frequency of knowledge
sharing. Even if a recipient is highly confident in his
or her own capabilities, if he or she does not believe
that critical knowledge and credible information will be
shared (i.e., lack of trust for sharing knowledge), this
may offset strong self-efficacy when it comes to set-
ting high personal performance goals. Conversely, the
positive relationship between self-efficacy and setting
high-performance goals may be strengthened when the
knowledge recipient trusts the sender. Based on the fore-
going, we predict:
Hypothesis 2 (H2). The degree to which the knowl-
edge recipient trusts the sender will moderate the rela-
tionship between self-efficacy and goal setting such that
recipients’ self-efficacy will be more strongly related
to setting high goals when the recipient has a greater
degree of trust in the knowledge provider.
The Interactive Influence of Shared Knowledge and
Self-Set Goal Level on Performance
Goal-setting theory has highlighted the way goals moti-
vate individuals to achieve higher levels of performance
through four primary mechanisms: direction, effort, per-
sistence, and task strategies (Locke and Latham 2002).
The last of these is especially critical when individuals
work on complex tasks, as goals motivate individuals
to discover, use, and apply task-relevant knowledge and
strategies toward goal achievement. When performance
requires applying knowledge from multiple sources to
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
76 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
complex decisions, knowledge recipients need to have
both access to others’ knowledge and be motivated to
apply that knowledge (Argote et al. 2003, Katz and Allen
1982, Zander and Kogut 1995). As noted above, one
reason for the failure to adequately capitalize on avail-
able knowledge inside organizations has been attributed
to lack of motivation (Menon and Pfeffer 2003).
Goal setting can be an important means of providing
the necessary motivation to actively seek out and use
available knowledge to achieve high levels of perfor-
mance. Setting very difficult goals (stretch goals) often
requires the use of new strategies that may be identi-
fied only through access to previously unknown informa-
tion, with colleagues and peers potentially constituting
one important available source (Thompson et al. 1997).
The use of goal setting for motivating use of avail-
able new knowledge has been mentioned by authors in
the macroknowledge management literature (e.g., Gupta
and Govindarajan 2000), who argue that improvements
in performance should be more likely when would-be
knowledge recipients both set high goals for their own
performance and have access to new forms of knowl-
edge from others in the organization. This argument has
not been advanced theoretically nor directly tested, how-
ever, as no goal-setting study to date has experimen-
tally studied knowledge sharing with peers as a means
of acquiring strategy information. Although knowledge
sharing and goal setting often have independent direct
influences on performance, the highest levels of perfor-
mance require knowledge recipients to set high goals for
their own performance and have access to performance-
relevant knowledge that others share. In other words, the
interactive combination of goal setting and knowledge
received likely has an important influence on perfor-
mance (Chesney and Locke 1991, Durham et al. 1997).
Other studies also suggest that an important contingency
to the goal-setting performance relationship is whether
individuals have access to the types of knowledge that
will enable them to create and apply task-appropriate
strategies (e.g., Audia et al. 1996, Earley and Perry 1987,
Latham and Baldes 1975). Based on the extant literature,
we predict:
Hypothesis 3 (H3). Shared knowledge and knowl-
edge recipients’ self-set goals will interact to influence
recipients’ performance such that recipients who receive
more knowledge and set higher goals for their own per-
formance will exhibit the highest levels of performance.
Method
Participants
Participants included 120 undergraduates enrolled in up-
per-level management courses at a large mid-Atlantic uni-
versity. Their average age was 21.76 years (SD =318).
Participants were randomly assigned unseen partners,
and the resulting 60 dyads were randomly assigned to
three incentive conditions. Participants could earn up to
$20 based on their performance in the simulation (i.e., a
merit-based pay system).
Task
Research participants worked on CELCOM 21 (Audia
et al. 2000), a computer-based interactive management
decision-making simulation. Participants worked in two-
person groups, physically separated from their partners
at computers in different rooms that were networked to
allow for communication via an instant messaging sys-
tem. The simulation required each participant to func-
tion in the role of a manager of a start-up unit of a
rapidly growing cellular phone company. Each individ-
ual was a manager of a unit and each pair of units
represented a division. Therefore, each individual was
considered both a unit manager and a division partner
with respect to his or her teammate. We will refer to
the focal individual in each dyad as the unit manager
and refer to his or her partner as the division partner.
Unit managers were asked to make a series of strate-
gic decisions over a period of eight years (each year
representing a single-decision period) to increase their
unit’s market share. Performance in each year built on
the performance of prior years (i.e., performance was
cumulative). Within each decision area, unit managers
undertook strategic actions each year and then obtained
performance feedback.
The chosen task was well suited for this study for
several reasons. The task provided measures of individ-
ual performance, kept unit managers regularly informed
of the financial impact of their strategic decisions, was
complex and highly involving, has been used to replicate
and complement field study findings (Audia et al. 2000),
was amenable to the manipulation of task-relevant infor-
mation, and simulated occurrences in a real industry that
allowed us to study knowledge in a grounded setting.
Just as would managers of different units within the
same division, each unit manager in the CELCOM 21
simulation worked at his or her own computer in a
separate location and made decisions regarding his or
her unit independently. Division performance was cal-
culated based on the average performance of the two
units that constituted the division (i.e., pooled interde-
pendence, Van deVen and Ferry 1978). Each unit man-
ager was provided with unique information that would
improve decision quality for capturing market share (i.e.,
performance). Moreover, each unit manager’s unique
information—if shared and applied—would enhance the
performance of his or her division partner. Thus, unit
managers were interdependent regarding the exchange
of information that might benefit their performance.
Incentive Pay Conditions
The design of the study included three incentive pay
conditions: individual, hybrid, and group. Unit managers
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
Organization Science 18(1), pp. 71–88, © 2007 INFORMS 77
were not in competition with their division partners in
any of the three conditions, nor were they in competition
with any other participants (they were informed of this at
the start of the experiment). Unit managers in each divi-
sion in the individual condition received a merit bonus
based 100% on their own unit’s performance, measured
by market share individually obtained by the end of the
simulation. Thus, each unit manager and his or her divi-
sion partner in this condition could receive a different
monetary bonus, which was based entirely on their indi-
vidual performance in obtaining market share for the
unit they directly managed.
Unit managers in each division in the hybrid condi-
tion received a merit bonus that was based half on their
own unit’s performance (final market share) and half on
the performance of the division (the average of the unit
manager and his or her division partner’s final market
share). Computationally, this results in 75% of the bonus
for unit managers in the hybrid condition as determined
by their own unit’s performance and 25% based on their
division partner’s performance (because the division per-
formance was based on the summed performance for
the two units). Thus, unit managers in this condition
also could receive differing amounts of bonus money,
however, part of the bonus was based on the division’s
performance.
Unit managers in the group condition received a bonus
based on the division’s performance. In other words,
50% of the bonus for unit managers in the group con-
dition was determined by their own unit’s performance
and 50% was based on their division partner’s perfor-
mance in managing their unit. Bonuses for all three
incentive conditions were tied to four levels of perfor-
mance: $5 for capturing 20% of the cellular market, $10
for 27% market share, $15 for 34% market share, and
$20 for capturing 40% or more market share. These four
performance levels were based on market share range
and means attained during piloting.
Procedure
The study was conducted in two rooms containing net-
worked computers. After reading introductory mate-
rial regarding the simulation, unit managers were given
a CELCOM 21 demonstration. Unit managers then
received an initial set of strategies that, if followed,
would allow them to perform the simulation with a mod-
erate level of success. They then worked on the simula-
tion for four decision periods as task training.
At the end of the practice period, unit managers were
informed of the incentive condition. Each unit manager
was also given one of two sets of complementary, unique
strategic information.1Each set contained four specific
strategies intended to improve performance. Although
each set of unique strategies helped capture market share
when consistently applied, the two sets of strategies used
together would lead to greater unit success. Thus, obtain-
ing the highest levels of final market share required man-
agers to share their unique strategies. All unit managers
were told that their division partners might have other
information about strategies for gaining market share.
Unit managers were also told that they were not in com-
petition with their division partners, and were shown
how to communicate with their partners via an instant
messaging system.
After participants demonstrated an understanding of
the messaging system, they completed the first four years
of CELCOM 21 and then completed goal and self-ef-
ficacy measures. Unit managers then completed the last
four years. Last, unit managers completed a postexper-
imental questionnaire that included the norms and trust
measures in addition to a manipulation check. These
data were collected before providing any partner perfor-
mance or final bonus calculation information.
Two trained, independent coders, blind to condition
and hypotheses, coded all phrases contained in the
60 sets of instant messaging communications between
unit managers and their division partners. The coders
assessed through consensus the information shared be-
tween members of the dyads using the two dimensions
described below. Therefore, a unique set of transcript
codes was associated with each unit manager. Later,
a third trained, independent coder, blind to condition
and hypotheses, used the coding system developed and
used by the first two coders. Our interrater reliability
and Cohen’s kappa statistics (provided below) reflect the
agreement of this third coder with the judgments of the
initial two coders.
Measures
Self-Efficacy. Self-efficacy in achieving market share
was measured by asking unit managers their level of
confidence (according to a 0–100 scale) in achieving
eight increasingly higher levels of market share by the
end of the simulation. Levels increased by 5% incre-
ments and ranged from 5% to 40% or higher market
share. This approach to assessing self-efficacy has been
used in prior research (e.g., Bandura 1997, Bandura and
Jourden 1991, Gist and Mitchell 1992, Lee and Bobko
1994, Locke et al. 1984). The measure was collected
after unit managers had completed Year 4 of the simu-
lation and had received feedback on their performance
(market share) for the first half of the simulation. The
scale reliability was 0.94.
Self-Set Goal Level. Goal level was measured by ask-
ing unit managers: “What is the lowest percent of market
share for your unit that you would be satisfied with get-
ting by Year 8?” This measure was based on Mento et al.
(1992). Self-set goal level was collected after unit man-
agers had completed Year 4 of the simulation and had
received feedback on their performance (market share)
for the first half of the simulation.2
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
78 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
Norms for Knowledge Sharing. A 10-item measure
was developed for this study to assess the degree to
which division partners established norms for sharing
knowledge (see the appendix). It should be noted that,
because norms are shared expectations of appropriate
behavior between members of a group (Hackman 1992),
we conceptualized norms for knowledge sharing as an
emergent phenomenon functioning at the dyadic level
(Kozlowski and Klein 2000). Therefore, norms were
measured post hoc, and we aggregated ratings to the
dyad level by averaging responses across division part-
ners. Participants responded on a seven-point Likert-
type scale ranging from “very little” to “to a very great
extent.” The internal consistency of the scale at the indi-
vidual level was 0.96. A one-way analysis of variance
(ANOVA) based on dyad membership was significant
(F=262; p<0001); the intraclass correlation coeffi-
cients were ICC(1) =039 and ICC(2) =059. The large
ICC(1) value indicates that a large portion of the vari-
ability across respondents was a function of dyad mem-
bership. The ICC(2) value, a measure of group relia-
bility, was less than what might be normally expected
based on the ICC(1) value, because ICC(2) values are
influenced by group size (Bliese 2000).
Trust. A five-item measure was developed for this
study (see the appendix) to assess the level of trust
that developed between each unit manager and his or
her division partner based on the task-specific defini-
tion of trust, which defines trust as the willingness of
a party to be vulnerable to the actions of another party
based on the expectation that the other will perform a
particular task-specific action (Mayer et al. 1995). Par-
ticipants responded on a seven-point Likert-type scale
ranging from “very little” to “to a very great extent.
Both conceptually and operationally, trust was consid-
ered at the individual level—in other words, the degree
of trust that each unit manager had in their division part-
ner (i.e., Dirks and Ferrin 2001, Ferrin and Dirks 2003).
The internal consistency estimate of the trust scale at the
individual level was 0.95.3
Shared Knowledge. This variable measured the extent
of knowledge shared by each unit manager’s division
partner. Two coders blind to condition and hypotheses
together coded all communications sent between a unit
manager and his or her division partner at the phrase
level, and a third coder provided an independent assess-
ment of the phrases for reliability purposes. The tallies
obtained from the first two coders were used to formu-
late the score for shared knowledge. The first aspect of
this score was the quantity of unique strategies shared
by each unit manager’s division partner in Years 1–4.
These strategies had been distributed prior to the start of
the game. The interrater reliability for this aspect (a cor-
relation measuring agreement of the third coder with the
initial two at the level of coded phrases) was 0.95 p <
0001.
The second aspect of this score was the quantity
of information about general performance, performance
numbers, and all other numbers shared by each unit
manager’s division partner. Information about general
performance included phrases such as, “I am doing bet-
ter this year than last.” Information about performance
numbers was more specific, such as, “14.3% market
share—Year 5.” The quantity of nonperformance num-
bers captured the extent to which each unit manager’s
division partner shared numerical information unrelated
to performance. For example, one unit manager wrote,
“I have 8 dealers in Market A.” The interrater reliabil-
ity for this aspect was 0.95 p < 0001. Cohen’s kappa,
using the formula for the two-category coding scheme
(which considers both aspects of the knowledge-sharing
score described above) to control for agreement that may
have been a result of chance (Bakeman and Gottman
1997), was 0.89.
The third aspect of the knowledge-sharing score
was the number of exchanges received from each unit
manager’s division partner. No reliability estimate was
needed for this measure, as it was a simple count. The
three aspects of shared knowledge described above were
distinct separate measures; the total composite yielded
a comprehensive measure of knowledge sharing. The
alpha of this composite is not relevant here, as the com-
ponents were all discrete elements of knowledge shar-
ing and not alternative measures of an identical concept.
Each unit manager’s division partner’s component scores
were standardized, averaged, and then had a constant of
10 added to yield an overall knowledge-shared score.
Performance. Each unit manager’s performance was
measured as their final market share. Because man-
agers had to receive and use available knowledge to per-
form well, performance is a direct consequence of each
unit manager’s receipt and use of knowledge from their
division partner.4All unit managers were instructed at
the beginning of the simulation that performance and
amount of bonus would be based on the market share
attained by the end of Year 8. Final market share scores
ranged from 0–54.6% (M=2185%, sd =1225%).
Task Ability. Task ability was considered as a con-
trol variable. Although none of the unit managers had
ever played CELCOM 21 before, task ability may have
enabled some to perform better. Therefore, task ability
was assessed using the market share that managers had
gained by the end of the practice period.
Manipulation Check
To ensure that unit managers understood the incentive
conditions, they were asked at the conclusion of the ex-
perimental session which of the three types of bonus
arrangements they worked under. Choices included:
(1) “My pay was determined 100% by my individual
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
Organization Science 18(1), pp. 71–88, © 2007 INFORMS 79
performance” (i.e., individual incentive), (2) “My pay
was determined 75% by my performance and 25% by
my division partner’s performance” (i.e., hybrid incen-
tive), or (3) “My pay was determined 50% by my indi-
vidual performance and 50% by my division partner’s
performance” (i.e., group incentive).
The Spearman correlation based on the ordinal-by-
ordinal comparison of the three conditions by three ma-
nipulation check categories was 0.61 (t=827, p<
0001). Unit managers seemed to have some difficulty in
distinguishing the hybrid condition from the group con-
dition; this may have been because of a lack of clarity in
the phrasing of the manipulation check questions. The
hybrid conditions had been presented to the unit man-
agers as a payoff matrix listing what levels of perfor-
mance for each unit manager and their division partner
had to be attained to reach the different bonus levels,
rather than as the 75–25 breakdown as indicated by the
manipulation check. To ensure that this did not affect our
results, all analyses were run first with the entire sam-
ple. Analyses at the individual level were then run with
only those subjects who had correctly identified their
bonus condition for the individual-level analyses; cross-
level analyses were run with only those dyads for whom
the knowledge sender gave the correct bonus condition
response. These additional analyses were performed so
as to ensure that some participants’ inabilities to report
the condition correctly were not attenuating any relation-
ships between study variables. Results from the samples
that correctly identified their condition were virtually
identical to those found for the overall sample; these
results are available upon request from the authors.
Analytic Strategy
As a preliminary analysis, a set of planned ordinal
comparisons was used to directly compare the impact
of the three types of incentive systems on knowledge
sharing. Because the variables in this study represent
different levels of analysis, we used multilevel data
analysis methods where appropriate. Knowledge shar-
ing occurred at the individual level of analysis (based
on communications from the knowledge provider to the
recipient). Yet, both incentive condition and norms for
knowledge sharing were operationalized as dyad-level
variables. This is because both members of a given
dyad operated under (were nested within) the same
incentive system. The emergent norms for knowledge
sharing were by definition a function of the dyad as
well. Thus, Hypotheses 1 involved variables at both
the individual (dependent variable, knowledge sharing)
and dyad level (independent variables/moderators, incen-
tives and norms for knowledge sharing). To appropri-
ately test these cross-level relationships, we used random
coefficient modeling (RCM), relying on Version 3.0 of
the Nonlinear and Mixed Effects (NLME) program for
S-PLUS and R (Pinheiro and Bates 2000). RCM is
useful for testing cross-level relationships because it pro-
vides a more appropriate estimate of standard errors than
other analytic methods when data are nested in groups
and assumptions of independence, therefore, are not war-
ranted (Bliese 2002, Raudenbush and Bryk 2002). The
analyses used here are also sometimes referred to as
hierarchical linear modeling (HLM). RCM techniques
are explained further by Bliese (2002), Hofmann (1997),
and Raudenbush and Bryk (2002).
We estimated two random coefficient models that pre-
dicted the degree to which providers shared knowledge
with their division partner. The first model was intended
to test for the possible direct effects of incentives and
norms for knowledge sharing on shared knowledge. We
then estimated a second model intended to test Hypoth-
esis 1, that norms for knowledge sharing moderate the
relationship between incentives and knowledge shared,
by considering the effects of the interaction between
the two Level 2 (dyad level) variables (incentives and
norms for knowledge sharing) on the Level 1 (individual
level) variable (knowledge shared). All variables exam-
ined in these RCM analyses were grand-mean centered;
this reduces multicollinearity and facilitates model esti-
mation when main effects and interactive terms are both
present (Raudenbush and Bryk 2002).5
Hypothesized relationships involving self-efficacy,
trust, self-set goals, knowledge sharing, and performance
(Hypotheses 2 and 3) were examined using hierarchi-
cal regression, as these variables were all at the indi-
vidual level. Because variables at the individual level
may also have been influenced by a lack of indepen-
dence caused by dyad membership, however, we uti-
lized Bliese and Ployhart’s (2002) ANOVA approach to
model testing. We compared the regression model testing
Hypotheses 2 and 3 with RCMs that allowed the inter-
cepts of the dyads to randomly vary (which helps con-
trol for nonindependence in the data, Bliese and Hanges
2001). We found that the RCMs were not significantly
better in predicting dependent variables than the regres-
sion model. Therefore, we used the standard regression
models.
Because we expected the interaction between knowl-
edge sharing and goal level in predicting performance
(Hypothesis 3) to be in an ordinal form with perfor-
mance being significantly higher only for unit man-
agers who set high goals and whose division partners
shared knowledge, we used Bobko’s (1986) method of
modeling ordinal interactions. This was appropriate for
two reasons. First, we predicted an ordinal interaction
a priori wherein only one group was hypothesized to
be significantly different from the others. Second, we
determined that the nonindependence in the data with
respect to the variables in question was inconsequen-
tial (above). As noted by Bobko (1986) and Strube and
Bobko (1989), when ordinal interactions are predicted
a priori, and thus only one group is hypothesized to be
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
80 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
Table 1 Descriptive Statistics and Intercorrelations
NMSD 1234567
1. Incentive condition 60 200 082 025∗∗ 033∗∗ 021015 015 026∗∗
2. Norms for knowledge 60 404 154 — 046∗∗ 035∗∗ 071∗∗ 022037∗∗
sharing
3. Knowledge shared 120 1000 100 — 019043∗∗ 019037∗∗
4. Self-efficacy 120 7461 2097 — 027060∗∗ 045∗∗
5. Trust 120 472 147 — 017 037∗∗
6. Self-set goal 120 1000 100 — 044∗∗
7. Performance 120 2185 1225 —
Notes. Incentive condition coded as 1 =team, 2 =hybrid, 3 =individual.
p<005, ∗∗p<001.
statistically different from the other three, a two-step
procedure is used. This models the interaction by first
testing three of the ordinal means for equality, then test-
ing the mean of those three against the remaining mean
using a planned comparison.
Results
Descriptive Statistics
Table 1 presents the individual-level descriptive statistics
and zero-order correlations. The correlation table does
not account for the fact that individual-level relationships
might be affected by the nonindependent nature of the
data; relationships and significance tests associated with
incentive condition and norms for knowledge sharing
should be viewed with caution until properly modeled
in the RCM analyses.
The Interactive Influence of Incentives and
Norms on Shared Knowledge
A set of planned, a priori ordinal comparisons was per-
formed on shared knowledge to test whether more inter-
dependent incentive conditions were positively related
to higher levels of shared knowledge. This analysis
revealed that incentive condition had a significant effect
on sharing knowledge such that those in the group
condition (M=1040, sd =108) shared significantly
more knowledge with their partner compared to those
in the hybrid incentive condition (M=984, sd =051;
p<001) and those in the individual incentive condition
(M=976, sd =046; p<0001). Those in the hybrid
and individual incentive conditions did not differ from
each other on their level of knowledge shared ns.We
consider these findings further below.
Table 2 provides results from the RCM analyses pre-
dicting knowledge shared by each unit manager’s divi-
sion partner. Tests for variation in the slopes of the
relationship between incentive condition and knowl-
edge sharing in the different dyads did reveal signifi-
cant variation among dyads (likelihood ratio of 11.63,
p<001), which suggests that dyad-level variables such
as norms for knowledge sharing may moderate the
incentive knowledge-sharing relationship. The top half
of Table 2 depicts the main effects model that tests for
the direct influence of incentive condition and norms for
knowledge sharing on the knowledge shared by divi-
sion partners. While we did not find incentive condi-
tion (where the three incentive conditions were treated
on a continuum) to be significantly knowledge shar-
ing as a direct effect in this analysis, note that the
planned contrast analyses reported above demonstrated
that group-based incentives were associated with sig-
nificantly more knowledge sharing than either hybrid-
or pure individually based incentives. The relation-
ship between norms and shared knowledge, however,
was significant (norms slope =020, t57 =383, p<
00). Additionally, this model accounted for 38.62%
of the total variance in knowledge shared. The lower
half of Table 2 shows strong support for Hypothesis
1; when individual-level and dyad-level norms, incen-
tive condition, and the interaction term were entered as
predictors of knowledge sharing, the interaction term
was significant (t56 =−280; p<001). This model
accounted for 50.08% of the total variance in knowledge
Table 2 Random Coefficient Model Results Predicting Knowl-
edge Shared by the Knowledge Sender
Parameter
Variable estimate SE df t-Test p
Main effects model
(Intercept) 1004 014 59 7424 000
Norms (individual level) 002 003 59 089 038
Incentive condition 013 010 57 130 020
(dyad level)
Norms (dyad level) 020 005 57 383 000
Interaction model
(Intercept) 1002 013 59 7459 000
Norms (individual level) 002 003 59 089 038
Incentive condition 014 009 56 153 013
(dyad level)
Norms (dyad level) 027 006 56 492 000
Condition ×norms 016 006 56 280 001
(dyad level)
Notes. Incentive condition coded as 1 =team, 2 =hybrid,
3=individual.
All variables examined in the RCM analyses described above
were grand-mean centered.
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
Organization Science 18(1), pp. 71–88, © 2007 INFORMS 81
Figure 2 Graphed Interactions
(a) Interaction between incentive condition and norms
on knowledge shared
(b) Interaction between self-efficacy and trust on
self-set goal level
(c) Interaction between self-set goals and knowledge
shared on performance
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
Low High
Self-set goal
Performance
Low knowledge shared
High knowledge shared
0.0
2.0
4.0
6.0
8.0
10.0
12.0
Individual Hybrid Team
Incentive condition
Knowledge shared
Low norms
High norms
Low trust
High trust
0.0
5.0
10.0
15.0
20.0
25.0
30.0
Low High
Self-efficacy
Self-set goal
sharing (R2for the interaction =011). The specific
form of the interaction is depicted in Figure 2(a) and
is consistent with our prediction that group-based incen-
tives would be more strongly related to knowledge shar-
ing when there were strong norms supporting knowledge
sharing.
The Interactive Influence of Self-Efficacy and Trust
on Recipient Goal Setting
Table 3 provides results from the hierarchical regression
analysis predicting the knowledge recipient’s self-set
goals. There is evidence of a main effect of self-
efficacy on self-set goals (=055; p<001), as would
be expected. More importantly, we found support for
Hypothesis 2, which predicted an interaction between
self-efficacy and trust in predicting goal level (=277;
p<001; R2=003). The interaction is plotted in
Figure 2(b). Consistent with our prediction, the rela-
tionship between knowledge recipients’ self-efficacy and
goal level was stronger when they trusted the knowledge
sender.
Table 3 Regression Results Predicting Knowledge Recipient
Self-Set Goal
Step 1 Step 2 Step 3
Variables   
Step 1
Practice performance 030∗∗ 020∗∗ 019
Step 2
Self-efficacy 055∗∗ 116+
Trust 005 171
Step 3
Self-efficacy ×trust 277∗∗
R2009 040 043
R2009 031 003
F1190∗∗ 2558∗∗∗ 2185∗∗∗
F 1190∗∗ 2954∗∗∗ 682
Notes. +p<010, p<005, ∗∗p<001, ∗∗∗p<0001.
The Interactive Influence of Shared Knowledge and
Self-Set Goals on Performance
Though not formally hypothesized, we did expect both
shared knowledge and self-set goals to have a direct
positive influence on performance. Using hierarchical
regression, we found that both goals and knowledge
shared by the division partner were significantly and
directly related to unit managers’ performance when
entered simultaneously in a regression equation after
controlling for task ability (=036, p<0001;
=029, p<0001 for goals and knowledge shared,
respectively; R2=023). Hypothesis 3 predicted an
ordinal interaction between knowledge shared and self-
set goal such that only those who set high goals and
whose partners shared knowledge would achieve a very
high level of market share. The a priori planned compar-
ison tests relating to the ordinal interaction demonstrated
the expected pattern of results such that the performance
of unit managers in the low knowledge shared/low goals
(M=1569; SD =888), low knowledge shared/high
goals (M=2118; SD =1056), and high knowledge
shared/low goals (M=2085; SD =1133) groups were
not statistically different from each other, but the mean
of the three groups (M=1924; SD =1027) was found
to be significantly lower than (t=517, p<001) the
high knowledge shared/high goals group (M=3182;
SD =1326). This interaction is graphed in Figure 2(c).
As expected, performance was highest when managers
set high goals for their performance and their division
partners shared knowledge.
Discussion
The overarching objective of this research was to move
closer to the development of an integrated model of
the motivational mechanisms that help explain and pre-
dict knowledge sharing and transfer between providers
and recipients. As pointed out through critiques of moti-
vation theory (Landy and Becker 1987, Pinder 1984,
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
82 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
Mitchell 1997), no single theoretical perspective can pro-
vide a complete explanation of the knowledge sharing
and transfer process. Different motivation theories may
be better at predicting specific aspects of the process,
such as the level of effort that providers demonstrate in
sharing their knowledge and the degree to which recip-
ients are motivated to apply new knowledge to improve
performance. Therefore, integrated theoretical perspec-
tives are required that combine existing motivational
theories to provide greater explanation of a specific phe-
nomenon than could be achieved by each theory applied
separately. Following this “middle-range” approach to
motivation theory development (Landy and Becker 1987,
Pinder 1984), we drew upon three distinct motivational
theories, each of which is well suited for understanding
a specific element of the interrelated set of motivational
processes at work in the interchange between knowl-
edge providers and recipient and in group performance.
What is novel about the perspective we have taken is that
(1) we combined these theories in complementary ways
by studying their interactions, rather than merely their
main or additive effects and (2) we applied the theories
to different parts of our model. Our findings yielded sev-
eral important insights and help to develop new theory
that can provide a more comprehensive explanation of
the knowledge sharing and transfer process.
Theoretical Implications
Our first theoretical contribution comes in the form
of demonstrating that knowledge sharing is better pre-
dicted by understanding how incentives and norms for
knowledge sharing interact rather than by examining
their impact separately. More specifically, we found that
incentives alone proved to have a rather weak influence
on knowledge sharing, but those effects were strength-
ened when mutual norms for knowledge sharing devel-
oped between the knowledge sender and recipient. It
is important to note that group norms had a minimal
effect on the relationship between incentives and knowl-
edge sharing for those working under individual-based
incentives. Hackman (1992) has suggested that norms
have an amplification quality in that they strengthen
the dominant messages being communicated. In our
study, norms made cooperation most salient when incen-
tives were more group based. The findings here support
the theoretical position taken by knowledge manage-
ment practitioners that structural means to motivate
knowledge sharing, such as incentives, need to be sup-
ported and reinforced by cultural elements that empha-
size open exchange and reciprocation (Davenport and
Prusak 1998, O’Dell and Grayson 1998, Stevens 2000).
Our results also support researchers who suggest that
knowledge management, in fundamental respects, is a
social process (Alavi and Leidner 2001, Brown and
Duguid 1991, Goodman and Darr 1998, Holtshouse
1998, Jarvenpaa et al. 1998, Levin and Cross 2004,
Szulanski et al. 2004).
Our second contribution involves integrating social
cognitive and goal-setting theories to develop new in-
sights into the motivational mechanisms that affect how
recipients set performance goals that ultimately impact
knowledge utilization. Consistent with literature, which
has demonstrated linkages between efficacy, goal set-
ting, and performance (e.g., Durham et al. 1997, Locke
and Latham 1990, Mulvey and Klein 1998, Prussia and
Kinicki 1996, Weldon and Weingart 1993), we found
that recipients who had greater levels of self-efficacy
were more likely to set high goals for their own per-
formance. Of more interest, however, was the finding
that the degree of trust that recipients have in knowledge
providers interacts with self-efficacy to affect goal set-
ting in interdependent situations. Recipients who were
confident in their own ability to perform well on the
task were more likely to set high goals for their own
performance when they trusted their partner. One could
say that trust increases one’s confidence in being able
to set goals in line with one’s task-specific confidence.
This role of trust had not been identified in previous
studies (Ferrin and Dirks 2003, McEvily et al. 2003,
Szulanski et al. 2004). This finding is also consis-
tent with the social motivation theory perspective (Geen
1991) in suggesting that trust influences how people
interpret and/or evaluate information provided by others
(Dirks and Ferrin 2001, Jarvenpaa et al. 2004). Further-
more, these findings, together with the results showing
that norms for sharing knowledge reinforce group-based
incentives, highlight the theoretical importance of inte-
grating organizational and social factors when studying
knowledge sharing.
The third theoretical contribution of this study is in
demonstrating that successful knowledge transfer ulti-
mately requires heightened levels of motivation on both
the part of providers to actively and openly share what
they know and recipients to set goals that require them
to seek out and integrate new knowledge from providers.
Notably, while both knowledge sharing and goals had
significant direct effects on performance, individual per-
formance was highest only when knowledge sources
shared their knowledge with recipients and recipients
set high goals for their own performance. By bringing
goal-setting theory into the knowledge-sharing literature,
we showed for the first time that knowledge sharing
interacts with goals to affect performance. This finding
indicates that knowledge that is shared may not always
be fully applied unless accompanied by some motiva-
tional inducement on the part of recipients to use that
knowledge (Katz and Allen 1982, Zander and Kogut
1995). Although previous results showed that task strate-
gies were important for goal attainment, especially in
complex tasks, previous studies either had strategies pro-
vided by the experimenter or had subjects learn the
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
Organization Science 18(1), pp. 71–88, © 2007 INFORMS 83
strategies on their own. None had looked at knowledge
sharing between peers or team members as a source of
task knowledge.
Fourth, an important theoretical and methodological
contribution of our research is that we explicitly rec-
ognized the multilevel nature of the variables in our
theoretical framework. This contribution is particularly
important given that the knowledge-sharing literature
has noted the importance of individual, social, and
organizational-level issues in understanding motivation
(e.g., Goodman and Darr 1996, Szulanski 1996). Despite
this acknowledgment, multilevel issues are often ignored
in knowledge-sharing research, particularly when it
comes to operationalizing and testing multilevel rela-
tionships. Relationships among variables in organiza-
tions are often hierarchical in nature necessitating the
consideration of levels of analysis issues (Bliese 2002,
Hofmann 1997, Kozlowski and Klein 2000, Raudenbush
and Bryk 2002). Such was the case here, owing to the
fact that both members of a given dyad operated under
the same organizational incentive system. This nesting
violates assumptions of data independence and requires
analysis with a method, such as the random coefficient
modeling used here, which takes into account the hierar-
chical nature of the data. Similarly, the emergent norms
for knowledge sharing were by definition a function of
the dyad, and these data were also subject to cross-level
analysis via random coefficient modeling to more fully
understand how group norms affect knowledge shar-
ing. Our approach points the way for researchers to
give further consideration to multilevel issues as they
affect our understanding of knowledge-sharing issues in
organizations.
Practical Implications
Our study has several practical implications. To begin,
our findings suggest that motivating knowledge shar-
ing can be achieved through incentives that emphasize
group performance and are strongly reinforced through
clear norms for sharing. Such reinforcements might be
achieved through emphasizing knowledge sharing in per-
formance appraisals, recognizing and praising those who
go out of their way to actively share their know-how
with others, consistently highlighting common goals and
objectives that can link potential knowledge providers
and recipients, and other steps that provide ongoing re-
minders that the open sharing of knowledge is an impor-
tant value associated with the organization’s culture.
Another practical implication is based on our finding
that the extent to which knowledge recipients trusted
the individual on whom they were dependent for new
knowledge was an important factor in whether recip-
ients’ self-efficacy translated into distinct performance
goals. Specifically, our results suggest that organizations
should take care to both develop the self-efficacy of
potential knowledge recipients (see Bandura 1997 for a
review regarding different ways to build self-efficacy)
and help to establish bonds of trust between interde-
pendent employees. Not much has been written about
how to build trust, but obviously it would be facilitated
by engaging in trustworthy actions such as being hon-
est, being open, and following through on commitments
(Jassawalla and Sashittal 1999). Such actions have to be
encouraged by a healthy moral climate and role model-
ing by senior managers, not to mention the fairness of
performance appraisal and related systems (Bartol and
Srivastava 2002, Ferrin and Dirks 2003).
Finally, it is also important for organizations to con-
sider the fact that encouraging potential senders to share
knowledge may not translate into a willingness to put to
use the knowledge that is received. Goal setting can be
one way to provide that motivation. Indeed, the under-
lying rationale behind the use of stretch targets is to
encourage individuals or teams to develop or acquire
new knowledge about alternative work strategies because
existing knowledge would be insufficient to achieve tar-
gets (Gupta and Govindarajan 2000, Thompson et al.
1997). Our findings suggest that organizations can help
create conditions for superior performance by setting
high-performance standards (e.g., through stretch goals)
and ensuring that those who are given high goals have
access to valuable outside knowledge.
Limitations and Directions for Future Research
This study does have a number of limitations that future
research might address. To begin, we used a labora-
tory simulation to achieve a high level of experimental
control and to measure actual knowledge sharing. Con-
sequently, we could not isolate all the conditions that
might occur in real organizational settings. For example,
the effects of norms and trust may have been under-
estimated, because the participants did not have estab-
lished working relationships. These effects may have
been stronger if examined in a field setting over a longer
time span, as group-level properties take time to develop
and may become a stronger influence on individual
behavior over time (e.g., Moreland and Levine 2001).
We also note that, given that both norms for knowl-
edge sharing and trust were necessarily measured after
assessing performance because they were emergent pro-
cesses, there may be a bias in the measure associated
with sense-making processes (see Staw 1975). Future
research might assess each concept in a more process-
oriented manner (e.g., by assessing these variables at
various points throughout the knowledge-sharing pro-
cess). Additionally, we point out that we did not directly
measure knowledge use on the part of the recipient, and
as a result, we cannot link the use of specific pieces of
knowledge to increases in performance. However, tran-
scripts of the conversations between knowledge senders
and recipients provide some qualitative evidence that
recipients were using knowledge provided by senders.
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
84 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
Furthermore, with motivational and social factors con-
trolled, knowledge sharing was significantly related to
performance; thus we can be certain that the knowledge
shared was actually used, i.e., performance reflected
knowledge utilization.
Another limitation is that the variables we examined
in this study undoubtedly have temporally dynamic rela-
tionships. Future studies might adopt a more explic-
itly process-oriented perspective to investigate questions
such as the conditions under which strong norms for
knowledge sharing and trust develop and how motiva-
tional mechanisms involved in knowledge sharing are
affected by violations of established norms and trust.
We also note that strategic decision making in an
organizational context may be more intense than was
captured here. Managers may face greater risks asso-
ciated with sharing versus withholding knowledge. In
some organizations, both sharing valuable knowledge
and admitting to needing information may be perceived
as risky (Borgatti and Cross 2003, McEvily et al. 2003).
Future research could examine the influence of risk lev-
els on the decision to share and use knowledge. This
perspective might extend to investigating the motiva-
tional processes involved in first seeking potential new
knowledge from various sources and then, subsequently,
weighing the risks associated with applying internally
derived knowledge versus utilizing know-how gleaned
from outside sources.
Although the task was engaging, the study’s partic-
ipants were undergraduate business students, not pro-
fessional managers. Therefore, future research might
investigate the conditions under which these findings
generalize to knowledge workers in actual organizational
settings. Still, we note that we have isolated major fea-
tures of a knowledge-sharing situation commonly found
in organizations through a carefully developed business
simulation. In addition, previous research has shown that
laboratory results replicate well in field settings (Locke
1986). Also, with respect to the task, participants essen-
tially completed an individual, noncompetitive decision-
making task with incomplete information (which could
then be supplemented by knowledge sharing). Questions
remain as to whether we would have found the same
results had the participants been working on a task that
involved joint decision making and/or some degree of
competition within or between dyads; future research
could explore these issues further.
This research focused exclusively on dyadic knowl-
edge sharing. In this way, we were able to directly
model motivational influences on knowledge providers
and recipients in the interest of building middle-range
motivation theory. Future research might productively
extend our research by focusing on the motivational
mechanisms influencing knowledge sharing within a
larger group or team context in which decisions are often
based on integrating different sources of knowledge. Our
findings might also be used to address how motivational
mechanisms affect the integration of knowledge among
team members, including further exploration of concepts
such as transactive memory systems and team mental
models (e.g., Faraj and Sproull 2000, Mohammed and
Dumville 2001, Moreland and Myaskovsky 2000). Addi-
tionally, it would also be helpful to study the motiva-
tion factors underlying knowledge sharing in distributed
teams, perhaps building on the work of Majchrzak et al.
(2005) relating to IT and collaboration know-how devel-
opment. Both group-level motivation mechanisms (e.g.,
collective efficacy) and decision making and knowledge
processing phenomenon (e.g., hidden profiles, Stasser
and Titus 2003) should be considered in future research.
Of interest, too, are other organizational practices and
policies, such as knowledge management systems and
procedures that govern communication, and other team-
level factors, such as task interdependence and goal
alignment, that work in combination across levels of
analysis to influence the motivation of potential knowl-
edge providers and recipients.
Our results also suggest that more work needs to be
done with respect to hybrid incentive systems. Consis-
tent with research by Ferrin and Dirks (personal commu-
nication, 2003), information sharing in the hybrid reward
condition (which involved providing conflicting incen-
tives for cooperation and competition simultaneously)
was more similar to information sharing in their indi-
vidual competitive condition than information sharing in
their cooperative condition. The fact that those working
under hybrid incentives failed to share more information
than those with only individual incentives suggests that
incentives aimed at stimulating knowledge sharing may
require a more significant or more clearly articulated
group-based component (Wageman 1995). Alternatively,
hybrid systems may be too complex for people to know
how to distribute their focus and energy (Bartol and
Durham 2000). Employees prefer to have at least some
component of incentive pay systems based on individual
contributions and performance (Kirkman and Shapiro
2000), so the question remains as to whether group and
individual systems can be successfully combined.
Conclusion
The results from this study add to the growing liter-
ature on knowledge sharing and utilization by provid-
ing insight into the motivational mechanisms associated
with both the knowledge provider and recipient. In par-
ticular, by integrating several existing theories of work
motivation and considering their interactive impacts, our
findings constitute a contribution toward the develop-
ment of a middle-range motivation-based theory (Landy
and Becker 1987, Pinder 1984) that explains knowledge
sharing and utilization. We hope that future research will
continue to utilize an integrative multilevel approach in
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
Organization Science 18(1), pp. 71–88, © 2007 INFORMS 85
seeking to develop a more comprehensive understanding
of the motivational complexities underlying knowledge
sharing, utilization, and, ultimately, performance.
Acknowledgments
The authors wish to thank Senior Editor Ann Majchrzak and
three anonymous reviewers for their helpful feedback and
suggestions. The authors also wish to thank Paul Bliese for
his generous advice regarding multilevel issues and analyses,
and gratefully acknowledge financial support from the Smith
School Strategic Research Fund, Robert H. Smith School of
Business, University of Maryland.
Appendix
Norms Supporting Knowledge Sharing
When working on the task, to what extent did it seem that you
and the other individual develop a mutual understanding that
each other would  (Scale: 1 =very little; 7 =to a very great
extent.)
1. share information on hints when you thought it might
help the other person.
2. share information on strategies that seemed to work
well.
3. let the other person know about what strategies or
decisions did not seem to work well.
4. go out of your way to help the other person with a
problem or question.
5. help the other person without being asked.
6. respond quickly to the other person’s e-mails.
7. send detailed messages in your e-mails to the other
person.
8. keep in touch with the other person during the task.
9. send the other person encouraging e-mail messages.
10. share information on market share achieved each
year.
Trust
To what extent did you personally  (Scale: 1 =very little;
7=to a very great extent.)
1. believe that the information offered by the other man-
ager would be useful to you.
2. believe that the information being provided by the
other manager was accurate.
3. trust the ability of the other manager to provide you
with useful information.
4. believe that the other manager was trying to help you.
5. trust the other manager to be honest with you when
communicating information.
Endnotes
1The four unique strategies given to unit managers can be
obtained from the authors upon request.
2Though we collected self-set goal levels both prior to the start
of the simulation and after Year 4, we elected to use the post-
Year-4 measures because they are more temporally proximal
to final performance (task ability was controlled in the anal-
ysis). In addition, we found that goal levels and self-efficacy
averaged across the two collection periods yielded very similar
results, except that the interactions we tested were less strong
(but still statistically significant).
3We conceptualized norms for knowledge sharing at the dyad
level and trust at the individual level. This theoretical distinc-
tion in levels of analysis reflects the fundamentally different
nature of the variables. However, because norms for knowl-
edge sharing and trust were highly correlated (r=071 at the
individual level), to ensure that these scales measured distinct
concepts, we compared the results of a one-factor to a two-
factor confirmatory factor analysis (each of which included the
10 items representing norms and the 5 items representing trust)
using LISREL VIII (Joreskog and Sorbom 1996). We found
that the two-factor model resulted in a significantly better
fit than the one-factor model (21=14990, p<0001);
model fit statistics for the two-factor model were 289=
19990; p=0001; CFI =089; NFI =082; NNFI =087).
4Based on the performance algorithm used by the simulation
as well as our own work piloting the simulation prior to the
start of data collection, we identified eight unique strategies
that, when implemented, directly impact performance. These
are the strategies, mentioned in Endnote 1, that were evenly
divided between manager dyads. There were no other alterna-
tive strategies that could have resulted in similarly high levels
of performance, and therefore the direct application of these
eight strategies was necessary to attain the highest levels of
performance.
5It is customary in random coefficient modeling analyses to
include both the individual and group level versions of vari-
ables that are conceptualized and aggregated to the group level
(cf. Bliese 2002, Hofmann 1997). Our results table therefore
includes both individual and dyad version of norms for knowl-
edge sharing. This allows one to control for any influence the
individual-level version of the variable in question may be
having on the dependent variable. We did not include an indi-
vidual version of the incentive condition variable, as each dyad
only was exposed to a single incentive condition—therefore,
there was no way to distinguish an individual-level effect.
References
Alavi, M., D. E. Leidner. 2001. Knowledge management and knowl-
edge management systems: Conceptual foundations and research
issues. MIS Quart. 1107–136.
Argote, L., P. Ingram. 2000. Knowledge transfer: A basis for compet-
itive advantage in firms. Organ. Behavior and Human Decision
Processes 82 150–169.
Argote, L., S. L. Beckman, D. Epple. 1990. The persistence and
transfer of learning in industrial settings. Management Sci. 36
140–154.
Argote, L., B. McEvily, R. Reagans. 2003. Managing knowledge in
organizations: An integrative framework and review of emerging
themes. Management Sci. 49 571–582.
Argote, L., P. Ingram, J. M. Levine, R. L. Moreland. 2000. Knowl-
edge transfer in organizations: Learning from the experience of
others. Organ. Behavior and Human Decision Processes 82 1–8.
Audia, P. G., E. A. Locke, K. G. Smith. 2000. The paradox of suc-
cess: An archival and a laboratory study of strategic persistence
following radical environmental change. Acad. Management J.
43 837–854.
Audia, P. G., A. Kristof-Brown, K. Brown, E. A. Locke. 1996. The
relationship of goals and microlevel work processes to perfor-
mance on a multipath manual task. J. Appl. Psych. 81 483–497.
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
86 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
Bakeman, R., J. M. Gottman. 1997. Observing Interaction: An Intro-
duction to Sequential Analysis, 2nd ed. Cambridge University
Press, Cambridge, UK.
Bandura, A. 1997. Self-Efficacy: The Exercise of Control.W.H.
Freeman & Company, New York.
Bandura, A., F. J. Jourden. 1991. Self-regulatory mechanisms gov-
erning the impact of social comparison on complex decision
making. J. Personality Soc. Psych. 60 941–956.
Barry, B., J. M. Crant. 2000. Dyadic communication relationships in
organizations: An attribution/expectancy approach. Organ. Sci.
11 648–664.
Bartol, K. M., C. C. Durham. 2000. Incentives: Theory and prac-
tice. C. L. Cooper, E. A. Locke, eds. Theory and Prac-
tice in Industrial-Organizational Psychology and Organizational
Behavior. Blackwell, Oxford, UK, 1–33.
Bartol, K. M., A. Srivastava. 2002. Encouraging knowledge shar-
ing: The role of organizational reward systems. J. Leadership &
Organ. Stud. 964–76.
Baum, J. A. C., P. Ingram. 1998. Survival-enhancing learning in
the Manhattan hotel industry, 1898–1980. Management Sci. 44
996–1016.
Bettenhausen, K. L., J. K. Murningham. 1985. The emergence
of norms in competitive decision-making groups. Admin. Sci.
Quart. 30 350–372.
Bliese, P. D. 2000. Within-group agreement, non-independence,
and reliability: Implications for data aggregation and analysis.
K. J. Klein, S. W. Kozlowski, eds. Multilevel Theory, Research,
and Methods in Organizations. Jossey-Bass, San Francisco, CA,
349–381.
Bliese, P. D. 2002. Multilevel random coefficient modeling in
organizational research: Examples using SAS and S-PLUS.
F. Drasgow, N. Schmitt, eds. Multilevel Theory, Research, and
Methods in Organizations: Advances in Measurement and Data
Analysis. Jossey-Bass, San Francisco, CA.
Bliese, P. D., P. J. Hanges. 2001. Non-independence in random coef-
ficient models: It is not only about too much Type I error. 61st
Acad. Management Meetings, Washington, D.C.
Bliese, P. D., R. Ployhart. 2002. Growth modeling using random coef-
ficient models: Model building, testing, and illustrations. Organ.
Res. Methods 5362–388.
Bobko, P. 1986. A solution to some dilemmas when testing hypothe-
ses about ordinal interactions. J. Appl. Psych. 71 323–326.
Bock, G. W., Y. Kim. 2002. Breaking the myths of rewards: An
exploratory study of attitudes about knowledge sharing. Inform.
Res. Management J. 15 14–21.
Borgatti, S. P., R. Cross. 2003. A relational view of information
seeking and learning in social networks. Management Sci. 49
432–446.
Brown, J. S., P. Duguid. 1991. Organizational learning and
communities-of-practice: Toward a unified view of working,
learning, and innovating. Organ. Sci. 240–57.
Brown, J. S., P. Duguid. 1998. Organizing knowledge. California
Management Rev. 40 90–111.
Brown, J. S., P. Duguid. 2001. Knowledge and organization: A social-
practice perspective. Organ. Sci. 12 198–213.
Brown, J. S., P. Duguid. 2002. Local knowledge: Innovation in the
networked age. Management Learn. 33 427–437.
Chesney, A. A., E. A. Locke. 1991. Relationships among goal diffi-
culty, business strategies, and performance on a complex man-
agement simulation task. Acad. Management J. 34 400–424.
Cohen, W. M., D. Levinthal. 1990. Absorptive capacity: A new per-
spective on learning and innovation. Admin. Sci. Quart. 35
128–153.
Cole, R. E. 1998. Introduction. California Management Rev.45(3)
15–21.
Cummings, L. L., P. Bromiley. 1996. The organization trust inventory
(OTI): Development and validation. R. M. Kramer, T. R. Tyler,
eds. Trust in Organizations: Frontiers of Theory and Research.
Sage Publications, Thousand Oaks, CA.
Davenport, T. H., L. Prusak. 1998. Working Knowledge: How Orga-
nizations Manage What They Know. Harvard Business School
Press, Cambridge, MA.
Dawes, R. M. 1980. Social dilemmas. Annual Rev. Psych.31
169–193.
Deutsch, M. 1949. An experimental study of the effects of compe-
tition and cooperation upon group process. Human Relations 2
129–152.
Deutsch, M., P. Gumpert, Y. Epstein. 1969. The effect of incentive
magnitude on cooperation the prisoner’s dilemma game. J. Per-
sonality Soc. Psych. 11 66–69.
Deutsch, M., Y. Epstein, D. Canavan, P. Gumpert. 1967. Strategies of
inducing cooperation: An experimental study. J. Conflict Reso-
lution 11 345–362.
Dirks, K. T. 1999. The effects of interpersonal trust on work group
performance. J. Appl. Psych. 84 445–455.
Dirks, K. T., D. L. Ferrin. 2001. The role of trust in organizational
settings. Organ. Sci. 12 450–467.
Durham, C. C., D. Knight, E. A. Locke. 1997. Effects of leader
role, team-set goal difficulty, efficacy, and tactics on team effec-
tiveness. Organ. Behavior and Human Decision Processes 72
203–231.
Earley, P. C., B. C. Perry. 1987. Work plan availability and perfor-
mance: An assessment of task strategy priming on subsequent
task completion. Organ. Behavior and Human Decision Pro-
cesses 39 279–302.
Faraj, S., L. Sproull. 2000. Coordinating expertise in software devel-
opment teams. Management Sci. 46 1554–1568.
Ferrin, D. L., K. T. Dirks. 2003. The use of rewards to increase
and decrease trust: Mediating processes and differential effects.
Organ. Sci. 14 18–31.
Geen, R. G. 1991. Social motivation. Ann. Rev. Psych. 42 377–399.
Gist, M. E., T. R. Mitchell. 1992. Self-efficacy: A theoretical analysis
of its determinants and malleability. Acad. Mgmt. Rev. 17 183–
202.
Goodman, P. S., E. D. Darr. 1996. Exchanging best practices through
computer-aided systems. Acad. Management Executive 10 7–19.
Goodman, P. S., E. D. Darr. 1998. Computer-aided systems and com-
munities: Mechanisms for organizational learning in distributed
environments. MIS Quart. 22 417–440.
Gupta, A. K., V. Govindarajan. 2000. Knowledge management’s
social dimension: Lessons from Nucor Steel. Sloan Management
Rev. 42(1) 71–80.
Hackman, J. R. 1987. The design of effective work teams. J. W.
Lorsch, ed. Handbook of Organizational Behavior. Prentice-
Hall, Englewood Cliffs, NJ.
Hackman, J. R. 1992. Group influences on individuals in organi-
zations. M. D. Dunnette, L. M. Hough, eds. Handbook of
Industrial and Organizational Psychology, 2nd ed. Consulting
Psychologists Press, Palo Alto, CA, 199–267.
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
Organization Science 18(1), pp. 71–88, © 2007 INFORMS 87
Hansen, M. T. 1999. The search-transfer problem: The role of weak
ties in sharing knowledge across organization subunits. Admin.
Sci. Quart. 44 82–111.
Hansen, M. T., M. L. Mors, B. Lovas. 2005. Knowledge sharing in
organizations: A relational perspective. Acad. Management J. 48
776–795.
Hayes, R. H., K. B. Clark. 1985. Explaining observed productivity dif-
ferentials between plants: Implications for operations research.
Interfaces 15 3–14.
Hofmann, D. A. 1997. An overview of the logic and rationale of
hierarchical linear models. J. Management 23 723–744.
Holtshouse, D. 1998. Knowledge research issues. California Manage-
ment Rev. 40 277–281.
Hornstein, H., M. Deutsch. 1967. The tendency to compete and attack
as a function of inspection, incentive, and available alternatives.
J. Personality Soc. Psych. 5311–318.
Jarvenpaa, S. L., D. Leidner. 1999. Communication and trust in global
virtual teams. Organ. Sci. 10 791–815.
Jarvenpaa, S. L., K. Knoll, D. E. Leidner. 1998. Is anybody out there?
Antecedents of trust in global virtual teams. J. Management
Inform. Systems 14 29–64.
Jarvenpaa, S. L., T. R. Shaw, D. S. Staples. 2004. Toward contextual-
ized theories of trust: The role of trust in global virtual teams.
Inform. Systems Res. 15 250–267.
Jassawalla, A. R., H. C. Sashittal. 1999. Building collaborative cross-
functional teams. Acad. Management Executive 12 50–63.
Joreskog, K. G., D. Sorbom. 1996. LISREL 8: User’s Reference
Guide. Scientific Software International, Chicago, IL.
Kalman, M. E., P. Monge, J. Fulk, R. Heino. 2002. Motivations to
resolve communication dilemmas in database-mediated collabo-
ration. Comm. Res. 29 125–154.
Kandel, E., E. P. Lazear. 1992. Peer pressure and partnerships. J. Polit-
ical Econom. 100 801–817.
Katz, R., T. J. Allen. 1982. Investigating the Not Invented Here (NIH)
syndrome: A new look at the performance, tenure, and commu-
nication patterns of 50 R&D project groups. R&D Management
12(1) 88–115.
Kirkman, B. L., D. L. Shapiro. 2000. Understanding why team
members won’t share: An examination of factors related to
employee receptivity to team-based rewards. Small Group Res.
31 175–210.
Kogut, B., U. Zander. 1996. What do firms do? Coordination, identity,
and learning. Organ. Sci. 7502–518.
Kostova, T. 1999. Transnational transfer of strategic organizational
practices: A contextual perspective. Acad. Management Rev. 24
308–324.
Kozlowski, S. W. J., K. J. Klein. 2000. A multilevel approach to
theory and research in organizations: Contextual, temporal, and
emergent processes. K. J. Klein, S. W. Kozlowski, eds. Multi-
level Theory, Research, and Methods in Organizations. Jossey-
Bass, San Francisco, CA, 3–90.
Landy, F., W. S. Becker. 1987. Motivation theory reconsidered. Res.
Organ. Behav. 91–38.
Latham, G. P., J. J. Baldes. 1975. The “practical significance” of
Locke’s theory of goal-setting. J. Appl. Psych. 60 122–124.
Lawler, E. E., III. 1981. Pay and Organizational Development.
Addison-Wesley, Reading, MA.
Lee, C., P. Bobko. 1994. Self-efficacy beliefs: Comparison of five
measures. J. Appl. Psych. 79 364–370.
Levin, D. Z., R. Cross. 2004. The strength of weak ties you can
trust: The mediating role of trust in effective knowledge transfer.
Management Sci. 50 1477–1490.
Locke, E. A. 1986. Generalizing from Laboratory to Field Settings.
Lexington Books, Lexington, MA.
Locke, E. A., G. P. Latham. 1990. A Theory of Goal-Setting and Task
Performance. Prentice-Hall, Englewood Cliffs, NJ.
Locke, E. A., G. P. Latham. 2002. Building a practically useful theory
of goal setting and task motivation. Amer. Psych. 57 705–715.
Locke, E. A., E. Frederick, C. Lee, P. Bobko. 1984. The effect of
self-efficacy, goals, and task strategies on task performance. J.
Appl. Psych. 69 241–252.
Majchrzak, A., A. Malhotra, R. John. 2005. Perceived individ-
ual collaboration know-how development through information
technology-enabled contextualization: Evidence from distributed
teams. Inform. Systems Res. 16 9–27.
Mayer, R. C., J. H. Davis, D. F. Schoorman. 1995. An integra-
tion model of organizational trust. Acad. Management Rev. 20
709–734.
McAllister, D. J. 1995. Affect- and cognition-based trust as founda-
tions for interpersonal cooperation in organizations. Acad. Man-
agement J. 38 24–59.
McEvily, B., V. Perrone, A. Zaheer. 2003. Trust as an organizing
principle. Organ. Sci. 14 91–103.
Menon, T., J. Pfeffer. 2003. Valuing internal vs. external knowledge:
Explaining the preference for outsiders. Management Sci. 49
497–514.
Mento, A. J., E. A. Locke, H. J. Klein. 1992. Relationship of goal
level to valence and instrumentality. J. Appl. Psych. 77 395–405.
Messick, D. M., M. B. Brewer. 1983. Solving social dilemmas:
A review. L. Wheeler, P. Shaver, eds. Review of Personality and
Social Psychology, Vol. 4. Sage Publications, Beverly Hills, CA,
11–44.
Miner, J. B. 2003. The rated important, scientific validity, and practi-
cal usefulness of organizational behavior theories: A quantitative
review. Acad. Management Learning Education 2250–263.
Mitchell, T. R. 1997. Matching motivational strategies with organiza-
tional contexts. Res. Organ. Behav. 19 57–149.
Mohammed, S., B. C. Dumville. 2001. Team mental models in
a team knowledge framework: Expanding theory and mea-
surement across disciplinary boundaries. J. Organ. Behav. 22
89–106.
Moreland, R. L., J. M. Levine. 2001. Socialization in organizations
and work groups. M. D. Dunnette, L. M. Hough, eds. Handbook
of Industrial and Organizational Psychology, 2nd ed. Consulting
Psychologists Press, Palo Alto, CA, 199–267.
Moreland, R. L., L. Myaskovsky. 2000. Exploring the performance
benefits of group training: Transactive memory or improved
communication? Organ. Behavior and Human Decision Pro-
cesses 82 117–133.
Mulvey, P. W., H. J. Klein. 1998. Impact of perceived loafing and col-
lective efficacy on group goal processes and group performance.
Organ. Behavior and Human Decision Processes 74 62–88.
Nonaka, I., H. Takeuchi. 1995. The Knowledge Creating Company.
Oxford, New York.
O’Dell, C., C. J. Grayson. 1998. If we only knew what we know:
Identification and transfer of internal best practices. California
Management Rev. 40 154–174.
Pentland, B. T. 1995. Grammatical models of organizational pro-
cesses. Organ. Sci. 6541–556.
Pinder, C. C. 1984. Work Motivation: Theory, Issues, and Applica-
tions. Scott, Glenville, IL.
Quigley, Tesluk, Locke, and Bartol: A Multilevel Investigation of the Motivational Mechanisms
88 Organization Science 18(1), pp. 71–88, © 2007 INFORMS
Pinheiro, J. C., D. M. Bates. 2000. Mixed-Effects Models in S and
S-PLUS. Springer-Verlag, New York.
Prussia, G. E., A. J. Kinicki. 1996. A motivational investigation of
group effectiveness using social-cognitive theory. J. Appl. Psych.
81 187–198.
Rau, D. 2005. The influence of relationship conflict and trust on the
transactive memory: Performance relationships in top manage-
ment teams. Small Group Res. 36 746–771.
Raudenbush, S. W., A. S. Bryk. 2002. Hierarchical Linear Models:
Applications and Data Analysis Methods, 2nd ed. Sage Publica-
tions, Newbury Park, CA.
Robinson, S. 1996. Trust and the breach of the psychological contract.
Admin. Sci. Quart. 41 574–599.
Roethlisberger, F. J., W. J. Dickson. 1939. Management and the
Worker. Harvard University Press, Cambridge, MA.
Rutte, C. G., H. A. M. Wilke. 1992. Goals, expectations, and behav-
ior in a social dilemma situation. W. Leibrand, D. Messick,
H. A. M. Wilke, eds. Social Dilemmas: Theoretical Issues and
Research Findings. Pergamon, New York, 280–305.
Spender, J. C., R. M. Grant. 1996. Knowledge and the firm: Overview.
Strategic Management J. 17 5–9.
Srinivas, T. 2000. A benchmarking method for business-process
reengineering and improvement. Internat. J. Flexible Manufac-
turing Systems 12 291–304.
Stasser, G., W. Titus. 2003. Hidden profiles: A brief history. Psych.
Inquiry 14 304–313.
Staw, B. M. 1975. Attribution of the “causes” of performance—
A general alternative interpretation of cross-sectional research
on organizations. Organ. Behavior and Human Decision Pro-
cesses 13 414–432.
Stevens, L. 2000. Incentives for sharing. Knowledge Management
(Oct.) 54–60.
Strube, M. J., P. Bobko. 1989. Testing hypothesis about ordinal inter-
actions: Simulations and further comments. J. Appl. Psych. 74
247–253.
Szulanski, G. 1996. Exploring internal stickiness: Impediments to the
transfer of best practice within the firm. Strategic Management J.
17 27–43.
Szulanski, G. 2000. The process of knowledge transfer: A diachronic
analysis of stickiness. Organ. Behavior and Human Decision
Processes 82 9–27.
Szulanski, G., R. Cappetta, R. J. Jensen. 2004. When and how trust-
worthiness matters: Knowledge transfer and moderating effect
of causal ambiguity. Organ. Sci. 15 600–613.
Thompson, K. R., W. A. Hochwarter, N. J. Mathys. 1997. Stretch tar-
gets: What makes them effective? Acad. Management Executive
11 48–60.
Tsai, W. 2001. Knowledge transfer in intraorganizational networks:
Effects of network position and absorptive capacity on busi-
ness innovation and performance. Acad. Management J. 44
996–1004.
Van de Ven, A. H., D. L. Ferry. 1978. Measuring and Assessing Orga-
nizations. John Wiley and Sons, New York.
Van Lange, P. A., W. Liebrand, D. Messick, H. Wilke. 1992.
Introduction and literature review. W. B. G. Liebrand, D. M.
Messick, H. Wilke, eds. Social Dilemmas: Theoretical Issues
and Research Findings. Pergamon, New York, 3–28.
Wageman, R. 1995. Interdependence and group effectiveness. Admin.
Sci. Quart. 40 145–180.
Wageman, R. 2001. The meaning of interdependence. M. E. Turner,
ed. Groups at Work: Theory and Research. Lawrence Erlbaum,
Mahwah, NJ, 197–218.
Wageman, R., G. P. Baker. 1997. Incentives and cooperation: The
joint effects of task and reward interdependence on group per-
formance. J. Organ. Behav. 18 139–158.
Weldon, E., L. R. Weingart. 1993. Group goals and group perfor-
mance. British J. Soc. Psych. 32 307–334.
Zaheer, A., B. McEvily, V. Perrone. 1998. Does trust matter? Explor-
ing the effects of interorganizational and interpersonal trust on
performance. Organ. Sci. 9141–160.
Zander, U., B. Kogut. 1995. Knowledge and the speed of the transfer
and imitation of organizational capabilities: An empirical test.
Organ. Sci. 676–92.
Zucker, L. G. 1987. Institutional theories of organization. Ann. Rev.
Soc. 13 443–464.
... Businesses are more effective when they can effectively establish conditions in which potential knowledge producers share their information and the recipients actively put it to use, despite the focus on and interest in motivating elements when investigating the information sharing and use process, the present literature has numerous shortcomings which means a flaw or inability to perform a given standard, usually in the character of people, a program, or a process. And in the case of emotional intelligence for example the person feels stressed out or sometimes emotional or stubborn or shy (Quigley et al., 2007). ...
... There are no cohesive, integrated theoretical frameworks of motivating variables that explain how information is transmitted between knowledge producers and users, and then used in ways that promote performance (Quigley et al., 2007), in particular, social motivation theory ideas such as trust have been utilized to assist explain knowledge transfer by academics interested in predicting it, for instance, academics have discovered that team-oriented incentive structures can promote collaboration, allowing companies to employ incentives to assist address the basic social problem that knowledge sender and receiver confront when sharing their expertise (Budur, 2018;Zaim et al., 2020). ...
... The sharing of information is a social practice is that norms and incentives work together to have a significant impact on knowledge-sharing behavior People in general have a need to interact with other people, which is known as social motivation, that has been critical to human existence, as humans do not do well on their own, for example there is an employee who is not doing his task correctly the manager should use the social motivation theory by putting this employee with another employee who is doing the task perfectly as a team to improve the other employees performance throughout motivation and as a result to that they will communicate with each other to share knowledge (Quigley et al., 2007). ...
Article
Full-text available
During the last few decades researchers has proven that controlling and sharing employee knowledge is critical for effective organizational transformation, innovation, and competitive advantage. Despite various efforts to promote information sharing inside businesses, individuals may not always be willing to share the knowledge assigned to them due to personal views or environmental restrictions that lead to knowledge concealment. We looked at the impact of emotional intelligence (EI) on knowledge sharing (KS) and knowledge hiding (KH). We collected 149 data points from various firms in Iraq to examine this effect. As a result, we discovered a positive impact of emotional intelligence and both knowledge sharing and concealment. employees who understand their feelings and other feelings they will tend to share their knowledge of the things that they want to share their knowledge and it will impact the company positively to increase creativity, innovation, teamwork, increasing productivity of the firm, but on the other hand, employees who do not fully understand their feelings and other feelings they will lean on hiding knowledge of their own and to conceal most parts. Therefore, in literature it will affect the company negatively (creating problems, lack of knowledge).
... Among them, the internal factors are the knowledge transfer subject and the own characteristic attributes of knowledge, and the external factors are the relevant environmental factors of knowledge transfer. Quigley et al. [9] identified team-oriented incentives, member self-efficacy, and self-goal setting and trusting relationship among members as the important factors affecting knowledge sharing efficiency. Luo et al. [10] investigated the co-evolution of complex networks and knowledge sharing based on a multi-intelligence model, and their simulation results showed that factors such as inter-subject knowledge distance, close association, and network cohesiveness had important effects on knowledge sharing efficiency. ...
Article
Full-text available
Knowledge transfer is the essential requirement for innovation clusters to carry out collaborative innovation, and it is a necessary process for innovation clusters to realize the knowledge value enhancement. The evaluation of knowledge transfer efficiency in innovation cluster can effectively reflect the knowledge gap, environment, and whether it is effectively coordinated among members of the innovation cluster. In order to evaluate the knowledge transfer efficiency in innovation clusters more scientifically and accurately, this paper analyzes the main factors affecting the efficiency of knowledge transfer based on the characteristics of innovation clusters and establishes a multi-level comprehensive evaluation system including knowledge transfer subject features, knowledge content features, knowledge transfer environment, and knowledge transfer coordination behavior. Furthermore, a set of AHP-Entropy index weight determination method and multi-level fuzzy comprehensive evaluation method are proposed to evaluate the knowledge transfer efficiency in innovation cluster. The results of the case study show that the evaluation system and method of knowledge transfer efficiency established in this paper are effective, and they can provide valuable reference for the management of knowledge transfer activities in innovation clusters.
... Job satisfaction concerns employees' evaluations of their jobs based on perceptions by comparing their actual job outcomes with desired ones (Schleicher et al., 2011). The concept is defined as a positive state where employees share their feelings about their job (Locke, 1976), ranging from moderate-to low-level satisfaction (Locke, 1976;Quigley et al., 2007). Furthermore, the idea is considered a causal factor that promotes intentions to stay with the organisation because it is a pleasant psychological state. ...
Article
Full-text available
Sketching on the Social Exchange Theory (SET), the present study aims to investigate the direct relationship between training and development, work environment, and job satisfaction with employee retention. The contingent role of transformational leadership was also analysed under the Situational Leadership Theory (SLT). Accordingly, we collected data from 287 employees of SMEs in northern China by employing a convenience sampling approach, exhibiting a response rate of 57.40 percent. The Partial Least Square-Structural Equation Modelling (PLS-SEM) analysis was then run to test the proposed hypotheses. The findings revealed a significant positive impact of training and development, work environment, and job satisfaction on employee retention. However, no moderating effect of transformational leadership was indicated on their direct relationship. This study has enriched the literature on employee retention and the leadership arena. To the best of the authors’ knowledge, there is no prior evidence concerning the study’s integrated relationship of the continuous variables. The implications and limitations were finally expressed at the end of this manuscript.
... These meetings offer opportunities to further understand the element's combinatorial possibilities, make their knowledge more accessible to others, and increase their influence (Andersson and Karlsson, 2007). Inventors who engage in the industry-or discipline-wide environment, on the other hand, can promote knowledge elements by increasing citations of their knowledge (Quigley et al., 2007). Higher increased citations bring more recognitions of the contributor's knowledge. ...
Article
Full-text available
This study conceptualises inventors as agents that connect knowledge elements within a knowledge network. Inventors’ searches thus focus on a knowledge element’s combinatorial potential–its suitability for combination with other knowledge elements to generate innovation. Using a survey-based network approach with scientists in the R&D department of a leading US oil and gas company, we find that internal search on combinatorial potential in a knowledge network has an inverted U-shaped relationship with innovation but external search on combinatorial potential has a positive relationship with individual innovation performance. No reinforcement effect between internal and external search was found. Instead, our study reveals the search strategy through knowledge networks that inventors are likely to reduce their effort on internal search but maintain external search to identify the combinatorial potential of knowledge elements, making them more likely to create fruitful knowledge combinations for innovation.
... De meest in het oog springende voordelen op korte termijn zijn dat deelnemers nieuwe vaardigheden en bekwaamheden aanleren, hun kennis up-to-date kunnen houden, en hun netwerk kunnen uitbreiden(Bouw et al., 2021a;Knol & Velzing, 2019;Noe, Clarke & Klein, 2014). Daarmee zijn zij beter in staat om te gaan met veranderende taakeisen en leveren zij betere werkprestaties(Quigley, Tesluk, Locke & Bartol, 2007). ...
Book
Full-text available
Grote maatschappelijke uitdagingen op het gebied van vergrijzing, duurzaamheid, digitalisering, segregatie en onderwijskwaliteit vragen om nieuwe manieren van werken, leren en innoveren. In toenemende mate wordt daarom ingezet op het bundelen van kennis en expertise van zowel publieke als private organisaties, die elkaar nodig hebben om te innoveren en complexe vraagstukken aan te pakken. Het concept ‘learning communities’ wordt gezien als dé oplossing om leren, werken en innoveren anders met elkaar te verbinden: collaboratief, co-creërend en contextrijk. Vanuit het Netwerk Learning Communities is een groep onafhankelijk onderzoekers van een groot aantal Nederlandse kennisinstellingen aan de slag gegaan met een kennissynthese rondom het concept ‘Learning Community’. Het Position paper is een eerste aanzet tot kennisbundeling. Een ‘levend document’ dat in de komende tijd verder aangevuld en verrijkt kan worden door onderzoekers, praktijkprofessionals en beleidsmakers.
Article
Purpose Past research on the motivational processes underpinning knowledge sharing has assumed that the sharing processes are similar for all individuals. Yet, sharing is a fundamental affiliative behavior, and the sharing processes can differ between people. This study aims to propose and test a model of the moderating influence that employee attachment patterns have on the theory of reasoned action (TRA)-defined knowledge sharing processes. Design/methodology/approach The authors administered a questionnaire to 1,103 employees from a range of industries who participated in an online Qualtrics survey. Advanced forms for structural equation modeling and latent profile analysis were used to assess the proposed model. Findings The results revealed that participants in the study exhibited the latent profiles corresponding to secure, dismissive, preoccupied and fearful patterns. The preoccupied cohort had the lowest knowledge sharing behavior, yet the strongest links within the sharing process. Secure, dismissive and fearful had similar sharing levels, but the strength of the TRA-defined processes differed. These findings underscore equifinality: although sharing may be approximately equal across different attachment patterns, the fundamental processes underpinning sharing differ. Research limitations/implications The authors used self-report data, given that sharing attitudes, norms and intentions may not be overly amenable to ratings even from well-acquainted others. Further, the use of advanced analytical methods helps to minimize common method concerns. Additionally, causal mechanisms underscoring the TRA have been demonstrated (Ajzen and Fishbein, 2005), allowing us to explore the moderating role of attachment patterns. Practical implications This study speaks to the importance of considering employees’ attachment patterns, and developing comprehensive intra-organizational norms, policies and systems that support and encourage knowledge sharing from employees with a variety of attachment patterns. Originality/value This study uniquely contributes to knowledge sharing literatures by incorporating attachment patterns as moderators within the TRA-defined sharing processes. The authors provide important insights on the role of individuals’ attachment patterns have for knowledge sharing behaviors, but also highlight how structure of knowledge sharing differed across subgroups of employees, determined based on their dispositional attachment pattern.
Thesis
Full-text available
The main focus of this study is twofold. Firstly, the thesis attempts to examine the direct relationship between individual (individual attitude, organizational commitment, and knowledge self-efficacy), organizational (organizational climate, management support, reward system and organizational structure), and interpersonal (interpersonal trust and social network) factors, and tacit knowledge sharing. Secondly, it is to examine the mediating effect of information and communication technology (ICT) usage on the relationship between individual, organizational and interpersonal factors, and tacit knowledge sharing. A total of 400 questionnaires were distributed to the technical staff of ICT sector in Jordan. Out of 400, only 375 questionnaires were returned. However, only 365 were usable for further analysis, representing a response rate of 92.75%. Hypotheses for direct relationships were tested using multiple regression, while the mediating effect were tested using the Preacher and Hayes analyses. Results indicated that individual attitude, knowledge self-efficacy, organizational climate, organizational structure, management support and interpersonal trust were significantly related to tacit knowledge sharing. However, the mediating analysis showed that ICT usage only partially mediated the relationship between knowledge self-efficacy, organizational climate, organizational structure and interpersonal trust, and tacit knowledge sharing. The current research have several implications. Knowledge wise, it provides understanding on the factors that affects tacit knowledge sharing. Practise wise, it suggests to managers that they should focus on providing the right management support, organizational structure and climate for sharing tacit knowledge. In addition, any practices that could promote organizational commitment, knowledge self-efficacy and interpersonal trust should also be implemented. The use of ICT should also be enforced so as to facilitate tacit knowledge sharing. Besides that, suggestions were also made for further research to be conducted the exploration of the variables tested in this study on other settings, and with different sample frames, in order to achieve a more robust finding towards a better understanding of the role of individual, interpersonal and organizational factors on tacit knowledge sharing. Keywords: Tacit Knowledge Sharing, Individual Factors, Organizational Factors. Interpersonal Factors, ICT Usage
Article
There is a growing trend of organizations deploying online user innovation communities (UIC) to collect innovative ideas from customers or users. When users submit a large number of ideas, screening and reviewing those ideas becomes a cumbersome task. Prior research indicates that, in a UIC, the popularity of an idea is one of the critical factors for getting recognition amidst a large number of ideas. There exists a lack of clarity on how an idea becomes popular in a UIC. Therefore, drawing on the theoretical underpinnings of cognitive overload theory, we develop a conceptual model to help firms understand how the text characteristics of an idea can be a determining force in making it popular. In particular, we study the effect of an idea's length, breadth, and textual dissimilarity to previously submitted ideas, on the likelihood of receiving comments and votes, which constitutes idea popularity. Our model is validated through logistic regression, using secondary data on 5283 users' ideas collected from the online UIC of the Starbucksmyidea platform. We find a significant impact of these characteristics on idea popularity. Implications for theory and practice are discussed for the effective functioning of UIC platforms.
Article
Full-text available
Cet article aborde la question des facteurs affectant le transfert de connaissances entre organisations. Cet article passe en revue la littérature existante afin de fournir un cadre d'analyse des déterminants qui peuvent affecter la performance de l'apprentissage. Nous abordons la question des partenariats en tant que transfert inter-organisationnel et discutons de trois types de déterminants liés aux partenaires, aux relations et aux connaissances associées. L'article propose un modèle théorique qui réunit les déterminants, l'efficacité de l'apprentissage et l'intérêt de l'entreprise partenaire sous la forme de la performance. Les entreprises doivent agir sur ces trois dimensions pour réussir leurs partenariats de transfert de connaissances et tirer pleinement parti de cette dynamique. L'article se termine par la construction d'un modèle théorique qui fera l'objet d'un test empirique ultérieur.
Conference Paper
Ample literature on social capital have provided inspiring results emphasizing the importance of social capital in various fields of development. The concept of Artificial Intelligence has become increasingly involved the area of computer science that highlights the creation of intelligent machines, which function and reaction like humans. Both Artificial Intelligence and knowledge management rotate around the same concept of knowledge where this connects with civic engagement, i.e., acting to produce a change in the communal life to develop new knowledge. This is a literature review critiquing the role of Artificial Intelligence and knowledge management within the realm of social capital of resources shared within the social media platforms for achieving civic engagement. The originality of this study originates from the scant model-based evidence which has yet to empirically assess what is the effect of social capital of knowledge management, and what is the effect of knowledge management on civic engagement. This is a research in progress bearing theoretical and practical implications.