Research Policy 42 (2013) 624–634

Contents lists available at SciVerse ScienceDirect

Research Policy
journal homepage: www.elsevier.com/locate/respol

Individual connectedness in innovation networks: On the role of individual

motivation夽
Rick Aalbers a,∗ , Wilfred Dolfsma b , Otto Koppius b
a
Radboud University, Institute for Management Research, Thomas van Aquinostraat 1, PO Box 9108, Nijmegen, The Netherlands
b
Rotterdam School of Management, Erasmus University, PO Box 1738, Rotterdam, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: Explanations of knowledge sharing in organizations emphasize either personality variables such as moti-
Received 14 March 2011 vation or network-related structural variables such as centrality. Little empirical research examines how
Received in revised form 15 October 2012 these two types of variables are in fact related: how do extrinsic and intrinsic motivation explain the
Accepted 23 October 2012
position that an employee entertains in a knowledge sharing network within an organization? Much is
Available online 20 January 2013
to be gained from a better understanding of how, empirically, psychological variables and an organiza-
tion’s network interrelate (Burt et al., 1998; Kalish and Robins, 2006; Moch, 1980; Teigland and Wasko,
Keywords:
2009). Still, this line of enquiry is not pursued much (Foss et al., 2009). This paper integrates the structural
Innovation network
Knowledge transfer
characteristics known to be implicated in knowledge transfer typically focused on in the social network
Motivation literature on the one hand, with the motivational perspective commonly identified in the organization
Centrality literature. This study examines how motivation – extrinsic (expected organizational rewards, recipro-
Inter-unit knowledge transfer cal benefits) and intrinsic (knowledge self-efficacy, enjoyment in helping others) – might explain how
employees may be better connected in the full knowledge transfer network or might be engaged more in
inter-unit knowledge transfer. Connectedness (closeness centrality) and inter-unit ties are well-known
to contribute to knowledge transfer. Analyzing data from a survey at two large European organizations,
this study, counterintuitively, shows that neither intrinsic nor extrinsic motivation explain an individual’s
favorable position in a knowledge transfer network.
© 2013 Published by Elsevier B.V.

1. Introduction 1996; Moorman and Miner, 1998; Kanter, 1985; Hargadon, 1998;
Perry-Smith and Shalley, 2003). Effectively orchestrating knowl-
As firms find themselves in increasingly competitive markets edge transfer to stimulate innovative outcomes certainly requires
and realize that they must be more innovative (Grant, 1996), further attention, however (Jackson et al., 2006).
the importance of knowledge transfer within their company is As pointed out by Foss (2007), organizations can seek to influ-
increasingly recognized. Knowledge may be spread throughout the ence individual actions to help accomplish favorable outcomes for
organization and not be available where it might best be put to use. the organization as a whole. Such orchestration may start with
Transfer of knowledge within the organization to gain competitive an understanding of both what motivates the individual to trans-
advantage has thus received considerable attention in the litera- fer knowledge, as well as, structurally, with whom individuals
ture (Grant, 1996; Teece et al., 1997; Moorman and Miner, 1998; exchange knowledge. The former is relevant to develop proper
Hansen, 1999). Scholars have emphasized that effective transfer HRM policy to stimulate knowledge transfer (Wasko and Faraj,
of knowledge between employees within an organization indeed 2000; Kankanhalli et al., 2005; Quigley et al., 2007). The latter is
increases the creativity and innovativeness of that same organiza- indicated by an individual’s position in the knowledge transfer
tion (Tushman, 1977; Ghoshal and Bartlett, 1988; Amabile et al., network of an organization (Wellman and Berkowitz, 1988; Diehl
and Stroebe, 1987; DeChurch and Marks, 2006). How each of these
contributes to knowledge transfer in a firm has been studied in
the past. Someone favorably positioned in the network in which
夽 We would like to thank editor and two anonymous reviewers for Research Policy innovative knowledge is transferred will perhaps unintentionally
as well as discussants and participants at the 2011 DRUID conference for suggestions, contribute more to firm-level outcomes than that of someone
criticisms and encouragement. not well-positioned (Obstfeld, 2005; Tsai, 2001). The relationship
∗ Corresponding author.
between network structure and individual motivation, however,
E-mail addresses: r.aalbers@fm.ru.nl (R. Aalbers),
w.a.dolfsma@rug.nl (W. Dolfsma), Okoppius@rsm.nl (O. Koppius). has not received much attention (exceptions are Kadsuhin, 2002;

0048-7333/$ – see front matter © 2013 Published by Elsevier B.V.

http://dx.doi.org/10.1016/j.respol.2012.10.007
 R. Aalbers et al. / Research Policy 42 (2013) 624–634 625

Kalish and Robins, 2006; Teigland and Wasko, 2009). Researchers the understanding of knowledge transfer within organizations and
commonly ignore how differences between individual might affect potentially benefits corporate innovation policies aimed at increas-
organizational network usage and benefits (Anderson, 20081 ). Nev- ing employee participation in knowledge transfer and innovation.
ertheless the structural patterns of relationships that emerge in
organizations and form the social infrastructure for the transfer of
innovative knowledge unavoidably implicate human psychology 2. Knowledge transfer within an organization:
(Moreno, 1961; Simmel, 1950). In this paper we study how individ- connectedness and motivation
ual motivation may explain an individual’s position in the structure
of the network in which innovative knowledge is transferred. As Finding the person within a multi-unit organization who pos-
such we aim to generate further insights into the psychology of the sesses the knowledge that one is looking for may be difficult
intra-organizational networks that facilitate the transfer of inno- (Szulanski, 2003; Hansen, 1999; Hansen and Haas, 2001). The rela-
vative knowledge. tive autonomy of units within a multi-unit organization structure
Although studies on the influence of intrinsic and extrinsic moti- can create a lack of awareness of each other’s activities on an
vation on individuals’ knowledge sharing behavior are not new (e.g. individual and a unit level, limiting knowledge-transfer. Within a
Amabile, 1993; Hung et al., 2011; Kankanhalli et al., 2005), scholars unit that specializes in one knowledge field, knowledge may also
have only begun to explore the effect of individual psychological be of the tacit kind. The advantage of the tacit nature of knowl-
differences on network structures (Klein et al., 2004; Anderson, edge is that imitation by competitors is relatively difficult (Nonaka
2008). The number of different issues addressed remains rather and Takeuchi, 1995), but at the same time the tacitness of the
limited which may in part be due to the difficulty of collecting rele- knowledge requires a high degree of personal contact to disperse it
vant data. The question as to how individual differences predispose throughout a company (Teece, 1998; Hansen, 1999). An individual’s
actors to position themselves in a network of relations thus has not capacity to contribute to the innovation processes in a firm then
received a persuasive answer as a result. Social network researchers depends not just on his own (absorptive) capacity originating from
seldom discuss the effects of individual psychological differences earlier experiences (Cohen and Levinthal, 1990), but also depends
on network structure and particularly not in the context of knowl- on the social, professional and hierarchical relations within the
edge transfer, while scholars in the field of HRM seldom consider organization.
social networks (Mehra et al., 2001; Minbaeva et al., 2003; Kaše If one is not well-connected one’s contribution to knowledge
et al., 2009). Although personality characteristics have occasion- transfer and thus the innovation process can be limited. There have
ally been linked to network position (a.o. Burt et al., 1998; Kalish been a number of recent calls to focus on the specific role of indi-
and Robins, 2006; Klein et al., 2004; Oh and Kilduff, 2008; Burt viduals in leveraging knowledge transfer (Felin and Hesterly, 2007).
et al., 2000), motivation has not been investigated in more detail While the literature on networks has been very helpful in suggest-
recently (with Foss et al., 2009 as a notable exception). This is ing the beneficial role of informal interpersonal ties for knowledge
somewhat surprising as the seminal work of a.o. Moreno (1961) transfer (e.g. Granovetter, 1973; Hansen, 1999), the actual process
and Simmel (1950) already emphasized the relevance of linking through which organizational knowledge is transferred remains
social structures and psychological processes. And even although relatively under-explored in the literature (Schulz, 2003; Reagans
more recent work has begun to link motivation to knowledge shar- and McEvily, 2003).
ing (a.o. Wasko and Faraj, 2000; Kankanhalli et al., 2005; Quigley In this paper we focus on the social network characteristics
et al., 2007), these studies continue to ignore the organizational known to particularly stimulate knowledge transfer within an orga-
network perspective. This study explicitly investigates the way in nization (Ibarra, 1993; Tsai, 2002; Nerkar and Paruchuri, 2005;
which motivation explains an individual’s connectedness in the Teigland and Wasko, 2009; Mäkelä and Brewster, 2009), and study
organizational knowledge transfer network, linking the structural how an individuals’ motivation helps explain how individuals will
patterns of relationships that emerge in organizations with human be well-positioned. More specifically we look at how an individual’s
psychology. motivation – extrinsic or intrinsic – explains their connectedness
We use the broadly accepted psychological construct of intrinsic in the organization-wide knowledge transfer network at large, and
and extrinsic motivation (Vallerand, 1997; Osterloh and Frey, 2000; also affects the maintenance in particular of inter-unit ties. Overall
Kankanhalli et al., 2005) to examine whether individuals with cer- connectedness and inter-unit ties in a knowledge transfer network
tain predispositions are indeed (1) better connected than others are each argued to be beneficial, as we explain below.
in a knowledge transfer network, in terms of closeness centrality, Individual motivation is indicated as the primary trigger for
or (2) more engaged in inter-unit knowledge transfer. Individ- knowledge transfer (Osterloh and Frey, 2000; Lin, 2007) and as key
uals that are well-connected within the full knowledge transfer determinant of successful or appropriate behavior by individuals
network of an organization, for instance, contribute significantly within organizations in general (Deci and Ryan, 1987). When an
more to beneficial outcomes including to innovative knowledge employee is motivated it means he/she is moved to do something,
transfer in particular (Nerkar and Paruchuri, 2005). We use the con- which turns motivation in a main concern of any manager (Hung
cept of closeness centrality to indicate the individual’s position in et al., 2011; Wasko and Faraj, 2005). Several prior studies explored
the full knowledge transfer network, rather than merely observ- conceptual (Bartol and Srivastava, 2002; Damodaran and Olpher,
ing her immediate connections. The connections an individual has 2000) or qualitative approaches (Weir and Hutchings, 2005; Yang,
may be within the own unit, while also knowledge transferred 2004) to study the motivatives fundamental to knowledge sharing
from other units, crossing unit boundaries, is believed to con- behavior. Motivation is believed to positively influence the amount
tribute to innovation in an important way (Burt, 2004; Aalbers et al., of knowledge transferred (Gupta and Govindarajan, 2000; Tsang,
forthcoming). We thus also determine how individual motivation 2002), and conversely lack of motivation in accepting knowledge
might positively influence her tendency to be involved in inter- from others leads to ‘stickiness’ or difficulties in the transfer pro-
unit knowledge transfer. By relating network structure elements to cess (Szulanski, 1996). Motivation is central to learning and lack of
motivational variables, this paper thus contributes significantly to motivation can hinder knowledge transfer (Pérez-Nordtvedt et al.,
2008).
In line with Osterloh and Frey (2000; Vallerand, 2000; Lin, 2007)
we identify two broad classes of motivation – extrinsic and intrin-
1
See Ibarra (1992, 1995) for one of the few exceptions related to gender and race. sic motivation. Extrinsic motivation focuses on the goal-driven
626 R. Aalbers et al. / Research Policy 42 (2013) 624–634

reasons, e.g. rewards or benefits earned when performing an activ- well-connected either directly or indirectly, allows one to access
ity (Osterloh and Frey, 2000). Intrinsic motivation indicates the information and muster support (Bala and Goyal, 2000). Well-
pleasure and inherent satisfaction derived from a specific activity connected individuals in a network are more likely to contribute
(Deci, 1975). Both forms have been found to influence individual to the development of relevant knowledge (Sparrowe et al., 2001;
intentions regarding an activity as well as their actual behaviors Wasserman and Faust, 1994). Well-connected individuals receive
(Davis et al., 1992; Lin, 2007). As a result of their predispositions, information and insights from many others, of higher accuracy,
individuals may shape their immediate network environment by and are more innovative than individuals that are positioned
(failing to) establish relations (Mäkelä and Brewster, 2009; Argote less strategically (Aalbers et al., forthcoming; Brass, 1984; Dekker
and Ingram, 2000). et al., 2003; Ibarra, 1993). Well-connected individuals can col-
Sharing2 knowledge may be extrinsically motivated as the con- lect and spread existing information more rapidly, but can also
sequence of such behavior is expected to lead to benefits for recombine existing ideas and knowledge in a novel way thus
the employee initiating in this activity (Osterloh and Frey, 2000; being more creative (Burt, 2004; Sparrowe et al., 2001). The more
Kankanhalli et al., 2005). In case of extrinsic motivation the shar- individuals are in regular contact with one another, the more
ing of knowledge will continue as long as the expected benefits likely they are to develop a “habit of cooperation” and act col-
equal or exceed the cost of participating in the exchange. Conse- lectively (Marwell and Oliver, 1988; Wasko and Faraj, 2005, p.
quently when the benefits no longer exceed the costs involved, 41).
the exchange will stop (Kelly and Thibaut, 1978). Benefits of Diversity of input from one’s connections is particularly rele-
being involved in knowledge transfer comprise of receiving orga- vant when considering the innovative knowledge transfer network.
nizational recognition and rewards or the obligation of other Such diversity can come from two different sources. First of all, an
colleagues to reciprocate with knowledge transfer at some moment individual may be receiving diverse information since she is well-
in the future (Ko et al., 2005). Costs typically relate to effort, connected in the full network. Closeness centrality indicates the
such as time spent, mental effort, preparation and so on (Lin, ease with which an individual is likely to receive information orig-
2007). inating from the different subgroups in a full network, through the
Sundgren et al. (2005) observed that information sharing indirect connections she has. In a direct sense, an individual may
requires self-initiated activities to fully benefit from the avail- connect with others who are in a different department or unit, and
able pool of knowledge. Self-initiated activities are influential as thus offer diversity of input(Teigland and Wasko, 2009).
they are primarily driven by intrinsic motivation (e.g. Dhawan An individual’s motivation and connectedness in the full network
et al., 2002). Engaging in the exchange of knowledge for its own may be argued to be conceptually connected. A better empiri-
sake, or for the pleasure and satisfaction derived from the experi- cal understanding of the link may increase our understanding of
ence, is a common indication that one is intrinsically motivated intra-organizational knowledge transfer. Connectedness need not
(Deci, 1975; Lin, 2007). The sharing of knowledge can in itself per definition imply that an individual is directly connected to all
be fulfilling for employees as it increases their own knowledge other colleagues, however. She may be able to reach others indi-
level or degree of confidence in their ability to provide knowledge rectly. Katz (1964) observed that those who are well-connected
that is useful to the organization (Constant et al., 1996). Previ- into networks of relationships in a professional environment will
ous research has demonstrated that people actually enjoy helping be more likely to participate in decision making, and see clearly
others by sharing knowledge and experience without an imme- how they contribute to group performance. Teigland and Wasko
diate or material benefit for themselves (Baumeister, 1982). Such (2009) extended this notion to cooperation patterns in a multina-
intrinsic motivations have been found to explain human behav- tional corporation setting and found that individuals who maintain
ior in various contexts (Vallerand, 2000; Vallerand and Ratelle, more social relationships with their peers will be more vital in the
2002). overall knowledge flows across the organization (see also Nerkar
Research on creativity has found that people will be most cre- and Paruchuri, 2005).
ative when they are primarily intrinsically motivated, rather than Moch (1980), then, observes that intrinsically motivated indi-
extrinsically motivated by expected evaluation, surveillance, dic- viduals are more socially integrated. The degree to which an
tates from superiors, or the promise of rewards (Amabile, 1997; individual is favorably positioned in the knowledge transfer net-
Teigland and Wasko, 2009). Knowledge workers have been found to work, in particular, is expected to be driven by intrinsic motivation
tend to be highly intrinsically motivated and often value knowledge for a number of reasons. Someone who is intrinsically motivated to
generation for its own sake (Mudambi et al., 2007). Furthermore share knowledge is more likely to volunteer knowledge that might
intrinsic motivation is positively associated with creativity (e.g. be relevant for a co-worker. In response to a request for knowledge
Amabile et al., 1996; Woodman et al., 1993). It is reasonable to from her social environment, an intrinsically motivated individual
expect that intrinsic motivation will have the same positive effects will be more likely to provide knowledge above and beyond what
on knowledge sharing as it has on other learning activities (Bock is asked for as the sharing of knowledge in itself is perceived as
et al., 2005; Burgess, 2005; Foss et al., 2009; Quigley et al., 2007; fulfilling (Constant et al., 1996). Intrinsically motivated individuals
Vallerand and Bissonnette, 1992; Vansteenkiste et al., 2004). This will also be approached more often to provide knowledge because
is supported by scholars who have argued that intrinsic motiva- alters expect that no immediate quid pro quo is expected or nego-
tion promotes knowledge sharing (Cabrera et al., 2006; Lin, 2007; tiated for; they are trusted more (Burt, 2005; Hansen, 1999). In
Osterloh and Frey, 2000). Hence, building on the insights from this such a context, no immediate return to time and effort invested
literature, we suggest that employees who are intrinsically moti- in knowledge transfer is to be expected and economic payoffs are
vated are more likely to share knowledge (Lin, 2007). highly uncertain (Dolfsma et al., 2009). For the individual to be well-
Existing research has taken an individual’s connectedness as connected in the full network, she depends on others to maintain
one of the most eminent indicators of an individual’s position in their connections so the focal individual can reach others indirectly.
a network. Being well-connected indicates the ease with which In the context of studying the full innovative knowledge trans-
someone can connect with any other alter in a network. Being fer network in an organization, these thus are reasons to expect
intrinsically motivated individuals to be well-connected. Hence we
hypothesize that intrinsic motivation is a useful predictor of an
2
We use the terms knowledge sharing, knowledge exchange and knowledge individual’s connectedness in the full innovative knowledge trans-
transfer interchangeably throughout this paper. fer network:
 R. Aalbers et al. / Research Policy 42 (2013) 624–634 627

Proposition 1. The degree to which an individual is well-connected et al., forthcoming), in which the same two individuals connect
within the full innovative knowledge transfer network, is positively through different networks, are less likely to develop between indi-
influenced by their intrinsic motivation. viduals from different units. Reciprocity in exchange may be less
likely (cf. Ensign, 2009). Levels of trust may then be lower between
3. Knowledge transfer within an organization: inter-unit individuals from different units who interact. The result may be
relations and motivation that more uncertainty is involved in inter-unit knowledge trans-
fer when compared to intra-unit knowledge transfer. Interactions
Aside from the benefits to the individual employee of being between individuals from the same unit tend to have a higher
connected well in the full innovative knowledge transfer network, expectation of results, albeit that these results are more incremen-
indirectly, innovative knowledge sharing will benefit from diver- tal; interactions between individuals from different departments
sity of direct relations an individual maintains (Whelan et al., may yield a more radical result, but the chances of the result mate-
2011). Individuals maintaining a larger number of such diverse rializing can be (much) lower (Constant et al., 1996; Whelan et al.,
contacts outside one’s own unit allows her to contribute better 2011). A high risk, uncertain yield environment that characterizes
to the innovative capacity of the organization (Tsai, 2002; Perry- an innovation setting where inter-unit knowledge transfer with
Smith and Shalley, 2003). Spanning unit boundaries provides access relatively less well known others from across unit boundaries is
to diverse sources of knowledge to an individual and its orga- involved, is likely in particular to attract individuals motivated by
nizational unit and is critical for an individual’s innovativeness immediate personal returns to knowledge exchange, such as career
within an organization (Aalbers et al., forthcoming; Burt, 2004). progression, status or financial rewards, to engage in knowledge
Differentiating between inter- and intra-unit knowledge transfer is transfer (Osterloh and Frey, 2000; Kankanhalli et al., 2005; Lin,
common to social network studies and has provided some interest- 2007).
ing insights regarding social capital, value creation and innovation When employees are to be actively encouraged to establish
(Tsai and Ghoshal, 1998; Tsai, 2002; Paruchuri, 2010; Mäkelä and and maintain diverse, inter-unit ties, they may then need to be
Brewster, 2009). Participation in cross-functional activity by indi- stimulated by relating to their immediate personal and profes-
viduals, for instance, increases their access to alternative views on sional interests, which entails that they must be sensitive to that
a firm’s existing strategy, goals, interests, time horizon, core val- (Amabile, 1997). Based on the previous arguments, we propose that
ues and emotional tone (Floyd and Lane, 2000), but also extends the increased perceived uncertainty and costs involved in inter-unit
their complementary functional expertise. Exposure to conflict and knowledge transfer indicate why inter-unit knowledge transfer
discussion as a result of different needs, objectives and interests may in particular appeal to an individuals’ extrinsic motivation.
between differentiated organizational units and hierarchical levels
Proposition 2. The number of inter-unit ties an individual holds in
is believed to increase ambidexterity at the individual level (Mom
the full innovative knowledge transfer network is positively influenced
et al., 2007). Maintaining diverse relations, directly, holds various
by their extrinsic motivation.
benefits to the individual, including in relation to her contribution
to innovative knowledge transfer.
Employees are most likely to interact with others in their imme- 4. Method and data
diate surroundings, however. Interacting with others, beyond the
immediate contacts or beyond whom one would as a matter of 4.1. Organizational settings
course meet regularly is more costly. Establishing and maintain-
ing ties is costly (Buechel and Buskens, 2012). Investment in (the Recognizing the need of more empirical support for the
expansion of) one’s network might become uneconomic especially theoretical findings to underscore the importance of inter-unit
when already supporting many ties.3 Indeed, studies of network communication structures (Hansen and Haas, 2001), this paper
connectedness find that the value of each connection maintained draws upon empirical research collected at a two separate com-
decreases with its distance, while the costs of establishing and panies. We have collected data for the full network of individuals
maintaining them increases, ensuring that actors in general strive involved in transfer of innovative knowledge in two companies in
to connect with others at a short distance (Jackson and Wolinsky, very different industries to provide indication of the representa-
1996; Hummon, 2000; Doreian, 2006). These costs might surge tiveness and robustness of our findings. As innovative activity is
in particular when ties span unit boundaries (Tsai, 2000; Haas a discretionary or extra-role activity, the network of individuals
and Hansen, 2005). An effort must be made to arrange meetings involved with transfer of innovative knowledge in a company is
to establish or maintain a contact. In addition, an employee that likely to be much smaller than the total number of individuals
acts outside his immediate surrounding is likely to have a dif- employed. A minority of individuals in any firm tends to be trans-
ferent social or professional thought world that can be difficult ferring innovative knowledge (Albrecht and Hall, 1991; Albrecht
(costly) to relate to. Unit membership tends to come with a shared and Ropp, 1984). We have aimed to collected information from
knowledge base and operational routines that have been there and about all these individuals to have an understanding of the full
from inception and is likely to have developed since (Gulati and innovative knowledge transfer network. In social network analysis,
Puranam, 2009). The diversity of or cognitive distance between it is common to work with data for relatively small networks (e.g.
specialized knowledge developed in separate units is very likely Dholakia et al., 2004; Albrecht and Hall, 1991; Tichy et al., 1979),
to be (much) larger than within a unit (Nooteboom, 2000). In addi- and centrality measures based on such small scale or even sam-
tion, knowledge transfer across unit boundaries tends to involve pled network data have proven to be robust in network studies
others with whom one interacts relatively less frequently as a mat- (Costenbader and Valente, 2003).
ter of course and with whom one is less familiar. Multiplex ties, One company studied is a subsidiary of a European electron-
known to be beneficial to innovative knowledge transfer (Aalbers ics and engineering conglomerate (Alpha Company), the other is a
leading European financial service provider (Beta Company).
Alpha Company is a multinational electronics and engineering
3
company headquartered in Europe. We study the Dutch sub-
The benefit of being well-connected by being on the shortest path to others in the
network (having a low closeness score) and of having diverse inter-unit ties should
sidiary, which has been in operation since the late 19th century
therefore be analyzed while controlling for number of an individual’s immediate and employs some 4000 employees. Alpha Company is organized
ties, as we explain in the methods section. according to a unit structure with a high level of autonomy and
628 R. Aalbers et al. / Research Policy 42 (2013) 624–634

responsibility for the separate units and the units are organized been personally informed of the purpose of the study to increase
according to product-market segmentation. Recently, the company their level of cooperation. The final version of the survey was sent
shifted its strategic insights from offering specific products toward in three rounds in each of the companies. The names mentioned at
offering ‘total solutions’ to its customers. As the company now aims Alpha Company by this first round of respondents (9) formed the
at offering integrated and innovative solutions based on its tech- input of respondents for the second round (42), who named another
nical competencies that cross unit boundaries, this heightens the round of respondents. Closure was reached after this third round
relevance of internal knowledge exchange and the network that of surveying. The full network studied consists of 83 employees
facilitates it. The unit structure constitutes a natural membership partaking in the knowledge transfer network, with a joint num-
boundary (see Hansen, 1999), however, and it is therefore that ber of 122 individual innovative knowledge transfer ties. The final
employees, sorted by unit membership, form the object of analysis overall response rate at Alpha Company was 96%. Only 4% did not
in this study of inter-unit transfer of knowledge. The selection of respond to the first mailing and the later three reminder mail-
these units is carried-out based on the input gathered during sev- ings. Following an identical procedure a comparable response was
eral interviews with the new business development director and achieved at Beta Company, with an overall response of 93%. With
the business managers in the separate units. Through the new busi- 30 employees at Beta Company partaking in round one, which
ness development director the commitment of the unit directors named another 54 employees that together formed the second
was sought and secured. survey round, the total innovative knowledge transfer network at
Beta Company is one of Europe’s largest and most innovative Beta Company showed to comprise off 144 employees. This inno-
payment processors, leading the market for secure payments and vation community together maintained 381 individual innovative
card processing solutions. We study its headquarters. With an knowledge transfer ties.
annual processing volume of almost 7 billion payments and the The invitation to participate in the survey was distributed by
switching of 1.9 billion POS and ATM transactions, the company’s e-mail at each of the companies, accompanied by a personalized
market share within the Eurozone is well over 10%, employing 1500 cover letter introducing the project and the hyperlink to the online
employees; with the large majority based in its European head- survey to the respondent, signed by the senior new business devel-
quarters. Beta Company is characterized by a strong unit structure. opment manager to improve response rates. An online survey was
Again access was negotiated through the director of the new busi- chosen to reduce the time needed to complete the questionnaire,
ness development unit, operating directly under the supervision of thus improving response rates. We did not opt to fix the num-
the board of directors. ber of contacts throughout the survey by using a list of names
provided by management or to indicate a limit to the number of
4.2. Data collection process possible contacts a respondent could list (Friedman and Podolny,
1993). However, we did issue a guideline of naming six employees
To test the formulated propositions, data on the social relations to make sure that only the most important contacts per employee
within both companies are gathered on individuals involved in the were mentioned. To reduce ambiguity regarding the interpretation
innovative knowledge transfer network. We follow Farace et al. of the questions by the respondents, the network questions were
(1977) to define social networks as repetitive patterns of interac- formulated in the native language.
tion among members of an organization. Data on the individual
level of the innovative knowledge transfer network, hereafter
5. Variables
referred to as the innovation network, are collected using semi-
structured interviews with managers and other employees as well
For each of the employees partaking in the innovation net-
as by means of an ego centric network survey. The interviews
work we collected input for each of the variables. The innovative
served a two-fold purpose: first, to become familiar with the orga-
knowledge transfer network was measured by asking individual
nizational setting and thus gain input for the proper design of the
respondents with whom they initiate a discussion of new ideas,
network survey and second, to determine the appropriate response
innovations and improvements regarding products and services
group within the company. In social network studies the most prag-
their unit offered (Borgatti and Cross, 2003; Cross and Prusak, 2002;
matic approach in an organizational setting is believed to be the
Rogers and Kincaid, 1981; Stephenson and Krebs, 1993; Rodan,
survey methodology (Borgatti and Cross, 2003; Wasserman and
2010). Based on the network data gained via the ego centric survey,
Faust, 1994). This study uses snowball methodology as the basis
the dependent variables of closeness centrality and interunit ties
for this survey. Snowball sampling is especially useful when the
were calculated, using Ucinet 6.0 (Borgatti et al., 2002; Freeman,
population is not clear from the beginning (Marsden, 1990, 2002;
1979).
Wasserman and Faust, 1994), which is the case for both organiza-
tions studied here. Innovative concepts may arise from employees
who are not part of a cross-unit team set up to stimulate innova- 5.1. Dependent variables
tion, for instance, or it may arise from interactions not mandated
by management. Snowball sampling is based upon several rounds 5.1.1. Individual connectedness
of surveying or interviewing where the first round helps to deter- Individual connectedness in the full network was measured
mine who will be approached as a respondent in the second round, by means of individual closeness centrality (Teigland and Wasko,
and so on. The first round of snowball sampling can be totally at 2009; Costenbader and Valente, 2003; Freeman, 1979). Closeness
random but it can be also based on specific criteria (Rogers and centrality takes the structural position of actors in the whole net-
Kincaid, 1981). To reduce the risk of ‘isolates’, i.e. isolated persons work into account, and has been identified as one of the most
within the organization who do possess relevant knowledge to a important centrality measures in network analysis (Borgatti, 2005).
particular subject, but who are being left out by the study due to Not only because it is the most appropriate centrality measure to
the lack of accuracy of random sampling (Rogers and Kincaid, 1981), determine the structural position of actors in terms of one’s effec-
this study opted in a first round to target respondents selected in tiveness of contacting all others in the network (Borgatti, 2005;
conjunction with new business development management. Freeman, 1979), but also because closeness centrality has proven
The networks analyzed are egocentric networks, an approach to be a robust measure (Costenbader and Valente, 2003). Close-
commonly adopted for the purposes of this kind of research. The ness centrality measures how many steps on average it takes for an
survey was first tested on a small sample of respondents whom had individual to reach everyone else in the network. Individuals who
 R. Aalbers et al. / Research Policy 42 (2013) 624–634 629

have high closeness centrality measures can most efficiently make employee. We included tenure to control for the effect of time, as
contact with others in the network (Freeman, 1979; Costenbader relations tend to develop throughout the years. Gender and unit
and Valente, 2003, p. 298). The higher one’s closeness centrality, membership were added to control for group affiliation effects.
the better positioned the individual is in dispersing information to Number of ties per individual employee was included to control for
other employees (Wasserman and Faust, 1994). In this study close- the effect of individual network size and the corresponding costs
ness centrality is preferred to degree centrality, as it does not only involved in maintaining or possibly increasing the number of con-
take into account direct connections among units but also indirect nections (Buechel and Buskens, 2012; Tsai, 2000; Haas and Hansen,
connections (Teigland and Wasko, 2009). An individual’s closeness 2005). For our analysis of the connectedness in the full network
centrality is the inverse of an individual’s closeness score, which is (closeness centrality), we controlled for number of cross-unit ties,
calculated4 as the sum of graph-theoretic distances from all other and vice versa.
individuals in the network, where the distance from one individ-
ual to another is defined as the length (in links) of the shortest path
from one to the other (Freeman, 1979). Closeness is an inverse mea- 6. Results
sure of centrality, a larger value indicates a less central actor while a
smaller value indicates a more central actor. For this reason we nor- Since aggregating the data for the two firms in our study into
malize the centrality score, following Borgatti and Halgin (2011) by a single dataset is both methodologically as well as substantially
dividing raw closeness by its maximum score in the database and meaningless, we provide analyses for each of the firms sepa-
extract this score from 1, which simultaneously reverses the mea- rately. Descriptives are presented in Tables 1a and 1b show the
sure so that high scores indicate greater connectedness. This allows means, standard deviations, and zero-order correlations of each
for easier interpretability of the results as well. Assuming that what of the variables for each company. Moving beyond these zero-
knowledge flows in a network originates from all other nodes with order results, the multiple regression analyses in Tables 2 and 3
equal probability and travels along the shortest path, highly cen- present the findings with regard to our first and second propo-
tral individuals have short distances from others, and so will tend sition, for each company. To make sure that the sample size did
to receive innovative information flows sooner (Borgatti, 2005, p. not lead to a violation of the normality assumption central to the
59). ordinary least square (OLS) procedure we used, we checked for
non-normal distributions and examined the skewness and kur-
5.1.2. Number of inter-unit ties
tosis of all the variables. The skewness and kurtosis showed no
The number of inter-unit ties was calculated based on data from
values greater than an absolute value of one (1) for each vari-
the ego-centric network survey. This variable was constructed from
able, suggesting reasonably normal distributions. Histograms for
the number of ties outside the unit, but inside the boundaries of
each variable were also examined, however, and these showed
the organization, that the individual employee maintained in the
that most scales were moderately positively skewed, with floor
previous three months (Tsai, 2000). We normalized this measure
effects evident for number of inter unit ties which appeared to
by dividing each individual score by the maximum in the dataset.
violate the assumption of normality. Thus a square root transfor-
mation was computed. The regression analyses were conducted
5.2. Independent variables
using both the non-transformed and transformed scores and this
was not found to make a statistically significant difference to
The independent variables intrinsic and extrinsic motivation were
the variance explained or to the regression coefficients. For sim-
derived from the Work Preference inventory of Amabile (1994).
plicity and interpretability of the findings reported upon, only
The Work Preference Inventory (WPI) is specifically designed to
the non-transformed scores are presented. Homoscedasticity was
assess individual differences in intrinsic and extrinsic motiva-
examined via several scatterplots and these indicated reasonable
tional orientations (1994). The questions of the inventory are
consistency of spread through the distributions. Multiple linear
specifically aimed to assess the major elements of intrinsic motiva-
regression analysis was deployed to determine which of the moti-
tion (self-determination, competence, task involvement, curiosity,
vational attributes predicts connectedness (closeness centrality)
enjoyment, and interest) and extrinsic motivation (concerns with
and number of inter-unit ties per employee in the knowledge trans-
competition, evaluation, recognition, money or other tangible
fer network.
incentives, and constraint by others). Drawing from a total reposi-
The results of the multiple regression analyses, presented in
tory of 30 propositions, Amabile points out that to fit the context of
Table 2, are remarkable. After running the model with the con-
the study we should match our findings accordingly. In this study
trol variables in isolation and after controlling for the specific
we draw from 6 propositions on intrinsic motivation and 6 proposi-
effect of number of ties as a proxy of an individual’s economic
tions on extrinsic motivation. These propositions were converted in
investments into his social infrastructure, models A3 and B3 intro-
12 questions for the questionnaire, framed on 7 point Likert scales.
duce intrinsic motivation. The inclusion of intrinsic motivation
The Cronbach alpha for the intrinsic motivation questions was .62,
in explaining individual connectedness results in a significant
the Cronbach alpha for the extrinsic motivation questions was .58.
improvement to the regression model at Beta Company (Model
For 33 percent of our respondents we were able to collect moti-
B3; F-test for R2 = 4.645, p < .05), identifying the relationship as
vational data on both intrinsic as well as extrinsic motivational
significant (Model B3; beta = −.278, p < .05). The sign for the effect
antecedents.
found in the case of Alpha Company is actually opposite to the
one found for Beta Company; the effects found for Beta Company
5.3. Control variables
are not statistically significant, however. In models A4 and B4 we
introduce extrinsic motivation as well. An individual’s motivation
Four variables were included as controls: tenure (in months),
is not a dichotomous matter, as we argued above, but might very
gender, unit membership, and number of ties per individual
well be based on a combination of both intrinsic and extrinsic
motives. The introduction of extrinsic motivation does not provide
4
a statistically significant beta and, in line with that, does not signif-
Closeness of a node is equal to the total distance (in the graph) of this node from
all other nodes. As a mathematical formula closeness, c(i), of node i can be written
icantly improve our model B4 results for Beta Company as a whole
as: c(i) = i dij where dij is the number of links in a shortest path from node i to
compared with model B3 (Table 2). A significant positive relation-
node j. ship between extrinsic motivation and connectedness, however,
630 R. Aalbers et al. / Research Policy 42 (2013) 624–634

Table 1a
Descriptive statistics Alpha Company.

Variable Means, standard deviations and correlations

Mean Std. dev. 1 2 3 4 5 6 7

1 Gender .925 .267

2 Tenure 10.666 6.325 .099
3 Unit 2.222 1.251 −.064 .078
4 Ties (#) 4.810 3.680 .26 −.087 −.083
5 Closeness centrality .127 .175 −.692** −.27 −.045 −.182
6 Intrinsic motivation 3.735 .481 −.059 −.233 .07 .087 .235
7 Extrinsic motivation 2.957 .516 .302 .288 .214 .181 −.564** .124
8 Inter-Unit ties 1.370 2.151 .117 −.05 .083 .636** −.05 .08 .246

N = 28.
* A significance level of 5%.
**
A significance level of 1%.
*** A significance level of .1%.

Table 1b
Descriptive statistics Beta Company.

Variable Means, standard deviations and correlations

Mean Std. dev. 1 2 3 4 5 6 7

1 Gender .793 .409

2 Tenure 7.450 4.654 −.20
3 Unit 2.31 1.547 .20 .01
4 Ties (#) 10.43 6.754 .06 −.44 −.11
5 Closeness centrality .145 .229 −.03 .22* .14 −.31***
6 Intrinsic motivation 5.155 1.105 .24* −.11 −.02 .04 −.29***
7 Extrinsic motivation 4.270 1.246 .05 −.02 .09 −.18 −.11 .19
8 Inter-unit ties 3.590 3.656 −.28* .15 −.14 .61*** −.16 −.03 −.23*

N = 58.
*
A significance level of 5%.
** A significance level of 1%.
***
A significance level of .1%.

does show for Alpha Company (Model A4; beta = −.419, p < .01). We of knowledge. As this effect appears to be limited to Alpha Com-
nonetheless, conservatively, interpret these findings as indicating pany only, we refrain from further speculation on the causes of
that Proposition 1 cannot be supported. this apparent relationship. What is more striking is the lack of sig-
The role of motivation for determining connectedness of indi- nificance for the control variable Tenure: one would expect that
viduals in a knowledge transfer network seems to be somewhat individuals are more likely to have developed more relations as
different for the two companies involved, suggesting that contin- they have been employed at a firm for a longer period of time,
gent elements, to be researched in future papers, may be at play including relations with ‘distant’ colleagues. This is not the case. In
beyond the scope of current research on motivation and involve- addition, being well-embedded locally, by having a large number
ment in knowledge transfer. From among the control variables we of direct ties in the knowledge transfer network, does not make an
include, it is striking to see how women at Alpha Company are employee well connected indirectly, at the network level, by having
more likely to be located in the network close to potential sources a higher closeness centrality.

Table 2
Motivation and closeness centrality (connectednessa ) – Proposition 1 tested.

D.V. Closeness centralitya Alpha Company Closeness centralitya Beta Company

Model A1 Model A2 Model A3 Model A4 Model B1 Model B2 Model B3 Model B4

Controls
Tenure .197 .201 .163 .036 −.139 −.115 −.097 −.106
Unit .073 .075 .090 .003 −.215 −.104 −.087 −.074
Gender .678*** .669*** .660*** .551*** .012 .015 −.053 −.051
# ties .032 .47 .012 .250 .251 .278

IVs
Extrinsic motivation .419** .123
Intrinsic motivation −.168 −.245* .278* .254

N 28 28 28 28 58 58 58 58
F-value 8.495*** 6.117** 5.192** 7.195** 1.286 1.742 2.418* 2.157
R2 .526 .527 .553 .683 .067 .116 .189 .202
Adjusted R2 .464 .440 .446 .588 .015 .049 .111 .109
F-test for R2 .044 1.233 8.249** 2.968 4.645* .881

Standardized coefficients. Durbin Watson model A: 1.837, VIF < 1.34, tolerance > 74; Durbin Watson model B: 1.877, VIF < 1.31, tolerance > 75.
a
Connectedness is operationalized as normalized closeness centrality at the employee level (see Section 4).
*
p < .05.
**
p < .01.
***
p < .001.
 R. Aalbers et al. / Research Policy 42 (2013) 624–634 631

Table 3
Motivation and inter-unit ties – Proposition 2 tested.

D.V. Inter-unit ties Alpha Company Inter-unit ties Beta Company

Model A1 Model A2 Model A3 Model A4 Model B1 Model B2 Model B3 Model B4

Controls
Tenure −.070 .002 −.036 −.040 −.092 −.009 −.006 −.001
Unit .096 .134 .102 .103 .106 .478** .471*** .473***
Gender .130 −.046 −.082 −.084 −.242 −.232** −.230* −.244**
# ties .659**** .636*** .637** .838*** .825*** .823***

IVs
Extrinsic motivation .144 .147 −.057 −.068
Intrinsic motivation −.015 .059

N 28 28 28 28 58 58 58 58
F-value .210 4.59** 3.308* 2.628* 1.934*** 25.101*** 19.969*** 16.551***
R2 .027 .425 .441 .441 .097 .655 .658 .661
Adjusted R2 −.100 .320 .307 .273 .047 .628 .625 .621
F-test for R2 15.219*** .601 .007 85.522*** .462 .473

Standardized coefficients. Durbin Watson model A: 2.665, VIF < 1.35, tolerance > 88; Durbin Watson model B: 1.874, VIF < 1.30, tolerance > 76.
*
p < .05.
**
p < .01.
***
p < .001.

Our second proposition looks at what explains the number of extrinsic motives on connectedness in the overall network (close-
inter-unit ties an individual has in the knowledge transfer network. ness centrality) are rather mixed. We thus find no indication that
Inter-unit ties have been found in the past to contribute to innova- individual motivation – extrinsic or intrinsic – favorably influences
tion in particular. Table 3 reports results of the multiple regression an individual’s position in a network where innovative knowledge
analyses for the datasets. Contrary to expectation, neither intrinsic is transferred.
nor extrinsic motivation of individuals predicts their involvement Motivation to be involved in knowledge transfer is thus differ-
in knowledge transfer across unit boundaries.5 The third and fourth ent from motivation to position oneself favorably in the network
model that add the motivation variables in comparison to the base in which innovative knowledge is transferred. The number of ties
models 1 and 2 offer no significant improvement as judged by the already maintained seems to prevent an individual from having
F-test for R2 . Betas are non-significant for both types of motives cross-unit ties, as one would expect drawing on costs of commu-
and so Proposition 2 must certainly be rejected. nication considerations (cf. Reagans and McEvily, 2003). The same
Entered as a control in models A2 and B2 (Table 3), the sheer does not hold for being well-connected in the whole network, how-
number of ties seems to be the best predictor of the inter-unit ties ever, as evidenced by the findings for closeness centrality. Given
an individual maintains in the innovation networks at both com- that an individual is dependent on possibly distant others to main-
panies. Statistically, the relation remains significant in each of the tain a path for knowledge transfer to him to persist, the costs of
models where this variable is included. Gender negatively impacts communication is not (fully) born by her, even when the diversity
the number of inter-unit ties an individual has in a statistically sig- of input thus received can be larger than from immediate cross-unit
nificant way only for Beta Company. Also departmental affiliation ties one maintains oneself. How exactly communication costs are
appears to matter in explaining the maintenance of inter-unit ties at implicated in an explanation of positioning in a knowledge transfer
Beta Company only. Again, and again surprisingly, having enjoyed a network is left for further research. In addition, the mutually inter-
long tenure at a company does not lead an employee to have more dependent nature of motivations, actions and positions in a social
inter-unit ties. environment may need to be more explicitly incorporated in an
analysis in future research (cf. Teigland and Wasko, 2009). Includ-
7. Discussion and conclusion ing reciprocal benefits as an extrinsic motivator (Lin, 2007; Kowal
and Fortier, 1999) might not adequately recognize the interdepen-
Connectedness and inter-unit ties in the knowledge transfer dencies and socially embedded exchange or transfer of knowledge
network are both, separately, known to allow individuals to con- over time (Bouty, 2000; Ensign, 2009). It might be more important
tribute to innovation (Burt, 1992; Tsai, 2001). The literature on for partners in knowledge transfer to have valuable knowledge to
people’s willingness to exchange knowledge suggests that indi- exchange (so as to call in a return favor later) than what motivates
viduals’ motivation should also be expected to be an important them to exchange in the first place (Bouty, 2000; Ensign, 2009, e.g.
explanatory for one’s favorable position in an innovative knowl- at p. 103).
edge transfer network, but fails to investigate this important As evidenced from the differing findings for the control vari-
claim (Anderson, 2008). We empirically investigate the role of ables included, organizationally contingent factors may play an
intrinsic and extrinsic motivation on one’s overall connectedness important role (cf. Lin, 2007). The effect of the variables Gender
and one’s diversity of ties in the intra-organizational innovation (Tables 2 and 3) and of Unit membership (Table 3) differs substan-
network. Analyzing data for the full network where innovative tially by company. The extent to which organizational fault lines
knowledge is transferred between individuals in two separate inno- (Bezrukova et al., 2009) differ between organizations can thus be
vative companies we present unexpected results. Intrinsic nor significant, something that a cross-sectional study without a focus
extrinsic motivation plays a role in determining an individual’s on social networks would not necessarily notice. Exactly how this
cross-unit knowledge transfer. The effect of individuals’ intrinsic or additional psychological factor relates to social networks needs to
be studied in future research.
Our study thus contributes to the understanding of intra-
5
Analysis of contribution from motivation – extrinsic and intrinsic – explicitly
organizational transfer of innovative knowledge and may steer
limited to intra-unit knowledge transfer provides similar findings. away management attention from too much attention to
632 R. Aalbers et al. / Research Policy 42 (2013) 624–634

individuals’ motives to others avenues for intervention. Manage- Borgatti, S.P., Halgin, D.S., 2011. On network theory. Organization Science 22 (5),
ment interventions have been found to hurt firm innovativeness 1168–1181.
Bouty, I., 2000. Interpersonal and interaction influences on informal resource
(e.g. Mellahi and Wilkinson, 2008; Shah, 2000). Although manage- exchanges between R&D researchers across organizational boundaries.
rial interventions are daily routine especially in large companies, Academy of Management Journal 43 (1), 50–65.
and involve substantial allocation of resources, studies on the Brass, D., 1984. Being in the right place: a structural analysis of individual influence
in an organization. Administrative Science Quarterly 29, 518–539.
effectiveness of managerial interventions are few and far between Buechel, B., Buskens, V., 2012. The dynamics of closeness and betweenness. Journal
(Okhuysen and Bechky, 2009, p. 482). If and when management is of Mathematical Sociology, forthcoming.
keen to sustain the structural features of a knowledge transfer net- Burgess, D., 2005. What motivates employees to transfer knowledge outside their
work unit? Journal of Business Communication 42, 324–348.
work, it may not want to focus too much exclusively on individuals’
Burt, R.S., Jannotta, J.E., Mahoney, J.T., 1998. Personality Correlates of Structural
motivations our study suggests. Innovation policy may thus fruit- Holes. Social Networks 20, 63–87.
fully focus in particular on other individual characteristics such as Burt, R., Hogarth, R.M., Michaud, C., 2000. The social capital of French and American
managers. Organization Science 11, 123–147.
skills (cf. Kaše et al., 2009) or on routines to be established in a firm
Burt, R.S., 2004. Structural holes and good ideas. American Journal of Sociology 110
(Zollo and Winter, 2002; van Driel and Dolfsma, 2009). (2), 349–399.
While a main strength of this paper is the analysis of data of Burt, R.S., 1992. Structural Holes: The Social Structure of Competition. Harvard Uni-
actual knowledge transfer in multiple firms, so firmer conclusions versity Press, Cambridge, MA.
Burt, R.S., 2005. Brokerage and Closure: An Introduction to Social Capital. Oxford
may be drawn than those conducted previously under controlled University Press, Oxford.
setting of an experiment in which, e.g. students participate (Quigley Cabrera, A., Collins, W.C., Salgado, J.F., 2006. Determinants of individual engagement
et al., 2007), nevertheless the study also holds limitations. First, in knowledge sharing. International Journal of Human Resource Management
17, 245–264.
we are not in a position to offer firm evidence for the causality of Cross, R., Prusak, L., 2002. The people who make organizations go or stop. Harvard
the findings presented. We need to rely on theory for that to some Business Review 80, 104–112.
degree. Second, the conceptual framework is not inclusive of all the Cohen, W.M., Levinthal, D.A., 1990. Absorptive capacity: a new perspective on learn-
ing and innovation. Administrative Science Quarterly 35, 128–152.
possible variables that could be related to innovative knowledge Constant, D., Sproull, L., Kiesler, S., 1996. The kindness of strangers: the usefulness
transfer, such as corporate reward system or corporate culture. In of electronic weak ties for technical advice. Organization Science 7, 119–135.
our study, we have rather focused on factors that have conceptual Costenbader, E., Valente, T.W., 2003. The stability of centrality measures when
networks are sampled. Social Networks 25, 283–307.
meaning beyond the idiosyncrasies of the firms included.
Damodaran, L.A., Olpher, W., 2000. Barriers and facilitators to the use of
We do stress that the knowledge sharing conditions in this study knowledge management systems. Behaviour and Information Technology 19,
are strongly focused on the transfer of innovative knowledge. Few 405–413.
Davis, F.D., Bagozzi, R.P., Warshaw, P.R., 1992. Extrinsic and intrinsic motivation
previous studies have this focus, limiting the comparability of our
to use computers in the workplace. Journal of Applied Social Psychology 22,
findings, in part because of the difficulties in collecting this type of 1111–1132.
data. It might be that the role of motivation is different in actual DeChurch, L.A., Marks, M.A., 2006. Leadership in multiteam systems. Journal of
business settings compared to experimental conditions. Applied Psychology 91, 311–329.
Deci, E.L., 1975. Intrinsic Motivation. Plenum Press, New York.
Deci, E.L., Ryan, R.M., 1987. The support of autonomy and the control of behavior.
Journal of Personality and Social Psychology 53, 1024–1037.
References Dekker, D., Franses, P.H., Krackhardt, D., 2003. An equilibrium-correction model for
dynamic network data. Journal of Mathematical Sociology 27, 193–215.
Aalbers, H.L., Dolfsma, W.A., Leenders, R.Th.A.J. Vertical and horizontal cross-ties: Doreian, P., 2006. Actor network utilities and network evolution. Social Networks
benefits of cross-hierarchy and cross-unit ties for innovative project teams. 28 (2), 137–164.
Journal of Product Innovation Management, forthcoming. Dhawan, S.K., Roy, S., Kumar, S., 2002. Organizational energy: an empirical study in
Albrecht, T.L., Hall, B.J., 1991. Facilitating talk about new ideas: the role of per- Indian R&D laboratories. R&D Management 32, 397–408.
sonal relationships in organizational innovation. Communication Monographs Dholakia, U.M., Bagozzi, R.P., Klein Pearo, L., 2004. A social influence model of con-
58, 273–288. sumer participation in network- and small-group-based virtual communities.
Albrecht, T.L., Ropp, V.A., 1984. Communicating about innovation in networks of International Journal of Research in Marketing 21, 241–263.
three U.S. organizations. Journal of Communication 34, 78–91. Diehl, M., Stroebe, W., 1987. Productivity loss in brainstorming groups. Journal of
Anderson, M.H., 2008. Social networks and the cognitive motivation to realize net- Personality and Social Psychology 53, 497–509.
work opportunities: a study of managers’ information gathering behaviors. Dolfsma, W., Van der Eijk, R., Jolink, A., 2009. On a source of social capital: gift
Journal of Organizational Behavior 29, 51–78. exchange. Journal of Business Ethics 89, 315–329.
Amabile, T.M., 1993. Motivational synergy: toward new conceptualizations of intrin- Ensign, P.C., 2009. Knowledge Sharing Among Scientists. Palgrave Macmillan, New
sic and extrinsic motivation in the workplace. Human Resource Management York.
Review 3, 185–201. Farace, R.V., Monge, P.R., Russell, H.M., 1977. Communicating and Organizing.
Amabile, T.M., 1994. Recognizing creativity: a reply to Magyari-Beck. Creativity and Addison-Wesley, Reading, MA.
Innovation Management 3, 244–245. Felin, T., Hesterly, W.S., 2007. The knowledge-based view, nested heterogeneity, and
Amabile, T.M., 1997. Motivating creativity in organizations: on doing what you love new value creation: philosophical considerations on the locus of knowledge.
and loving what you do. California Management Review 40, 39–58. Academy of Management Review 32, 195–218.
Amabile, T., Conti, R., Coon, H., Lazenby, J., Herron, M., 1996. Assessing the work Floyd, S.W., Lane, P.J., 2000. Strategizing throughout the organization: manag-
environment for creativity. Academy of Management Journal 39, 1154–1184. ing role conflict in strategic renewal. Academy of Management Review 25,
Argote, L., Ingram, P., 2000. Knowledge transfer: a basis for competitive advantage 154–177.
in firms. Organizational Behavior and Human Decision Processes 82, 150–169. Foss, N.J., 2007. The emerging knowledge governance approach: challenges and
Bala, V., Goyal, S., 2000. A noncooperative model of network formation. Economet- characteristics. Organization 14, 29–52.
rica 68, 1181–1229. Foss, N.J., Minbaeva, D.B., Pedersen, T., Reinholt, M., 2009. Encouraging knowledge
Bartol, K., Srivastava, A., 2002. Encouraging knowledge sharing: the role of orga- sharing among employees: how job design matters. Human Resource Manage-
nizational reward systems. Journal of Leadership and Organization Studies 19, ment 48, 871–893.
64–76. Freeman, L.C., 1979. Centrality in social networks: conceptual clarification. Social
Baumeister, R.F., 1982. A self-presentational view of social phenomena. Psycholog- Networks 1, 215–239.
ical Bulletin 91, 3–26. Friedman, A.F., Podolny, J., 1993. Differentiation of boundary spanning roles: labor
Bezrukova, K., Jehn, K.A., Zanutto, E.L., Thatcher, S.M.B., 2009. Do workgroup fault- negotiations and implications for role conflict. Administrative Science Quarterly
lines help or hurt? A moderated model of faultlines, team identification, and 37, 28–47.
group performance. Organization Science 20, 35–50. Granovetter, M., 1973. The strength of weak ties. American Journal of Sociology 78,
Bock, G.W., Zmud, R.W., Kim, Y.G., Lee, J.N., 2005. Behavioral intention for- 1360–1380.
mation in knowledge sharing: examining the role of extrinsic motivators, Grant, R.M., 1996. Toward a knowledge-based theory of the firm. Strategic Manage-
social–psychological forces, and organizational climate. MIS Quarterly 29, ment Journal 17, 109–122.
87–111. Ghoshal, S., Bartlett, C.A., 1988. Creation, adoption and diffusion of innovations
Borgatti, S., Everett, M.G., Freeman, L.C., 2002. Ucinet 6 for Windows. Analytic Tech- by subsidiaries of multinational corporations. Journal of International Business
nologies, Harvard, Massachusetts. Studies 19, 365–388.
Borgatti, S.P., 2005. Centrality and network flow. Social Networks 27 (1), 55–71. Gulati, R., Puranam, P., 2009. Renewal through reorganization: The value of incon-
Borgatti, S., Cross, R., 2003. A relational view on information seeking and learning in sistencies between formal and informal organization. Organization Science 20
social networks. Management Science 49, 432–445. (2), 422–440.
 R. Aalbers et al. / Research Policy 42 (2013) 624–634 633

Gupta, A.K., Govindarajan, V., 2000. Knowledge flows within multinational corpo- Mudambi, R., Mudambi, S.M., Navarra, P., 2007. Global innovation in MNCs: the
rations. Strategic Management Journal 21, 473–496. effects of subsidiary self-determination and teamwork. Journal of Product Inno-
Hansen, M., 1999. The search-transfer problem: the role of weak ties in sharing vation Management 24, 442–455.
knowledge across organization subunits. Administrative Science Quarterly 44, Nonaka, I., Takeuchi, H., 1995. The Knowledge-Creating Company: How Japanese
82–111. Companies Create the Dynamics of Innovation. Oxford University Press, New
Hansen, M.T., Haas, M.R., 2001. Competing for attention in knowledge markets: York, NY.
electronic document dissemination in a management consulting company. Nerkar, A., Paruchuri, S., 2005. Evolution of RandD capabilities: role of knowledge
Administrative Science Quarterly 46, 1–28. networks within a firm. Management Science 51, 771–785.
Hargadon, A.B., 1998. Firms as knowledge brokers: lessons in pursuing continuous Nooteboom, B., 2000. Learning and Innovation in Organizations and Economies.
innovation. California Management Review 40, 209–227. Oxford University Press, Oxford.
Haas, M.R., Hansen, M.T., 2005. When using knowledge can hurt performance: Obstfeld, D., 2005. Social networks, the Tertius Iungens orientation, and involvement
the value of organizational capabilities in a management consulting company. in innovation. Administrative Science Quarterly 50, 100–130.
Strategic Management Journal 26 (1), 1–24. Oh, H., Kilduff, M., 2008. The ripple effect of personality on social structure: self-
Hung, S.Y., Durcikova, A., Lai, H.M., Lin, W.M., 2011. The influence of intrinsic and monitoring origins of network brokerage, Journal of. Applied Psychology 93,
extrinsic motivation on individuals’ knowledge sharing behavior. International 1155–1164.
Journal of Human–Computer Studies 69, 415–427. Okhuysen, G.A., Bechky, B.A., 2009. Coordination in organizations: an integrative
Hummon, N.P., 2000. Utility and dynamic social networks. Social Networks 22, perspective. Academy of Management Annals 3 (1), 463–502.
221–249. Osterloh, M., Frey, B.S., 2000. Motivation, knowledge transfer, and organizational
Ibarra, H., 1992. Homophily and differential returns: sex differences in network forms. Organizational science 11, 538–550.
structure and access in an advertising firm. Administrative Science Quarterly Paruchuri, S., 2010. Intraorganizational networks, interorganizational networks, and
37, 422–447. the impact of central inventors: a longitudinal study of pharmaceutical firms.
Ibarra, H., 1993. Network centrality, power and innovation involvement: determi- Organization Science 21, 63–80.
nants of technical and administrative roles. Academy of Management Journal Perry-Smith, J.E., Shalley, C.E., 2003. The social side of creativity: a static and
36, 471–501. dynamic social network perspective. Academy of Management Review 281,
Ibarra, H., 1995. Race, opportunity, and diversity of social circles in managerial 89–107.
networks. Academy of Management Journal 38, 673–703. Pérez-Nordtvedt, L., Kedia, B., Datta, D.K., Rasheed, A., 2008. Effectiveness and effi-
Jackson, S.E., Chuang, C.H., Harden, E.E., Jiang, Y., 2006. Toward developing human ciency of cross-border knowledge transfer: an empirical examination. Journal
resource management systems for knowledge-intensive teamwork. Personnel of Management Studies 45, 714–744.
and Human Resources Management 15, 17–70. Quigley, N., Tesluk, P.E., Bartol, K.M., Locke, E.A., 2007. A multilevel investigation of
Jackson, M., Wolinsky, A., 1996. A strategic model of social and economic networks. the motivational mechanisms underlying knowledge sharing and performance.
Journal of Economic Theory 71, 44–74. Organization Science 18, 71–88.
Kadsuhin, C., 2002. The motivational foundation of social networks. Social Networks Reagans, R., McEvily, B., 2003. Network structure and knowledge transfer: the effects
24, 77–91. of cohesion and range. Adminstrative Science Quarterly 48, 240–267.
Kankanhalli, A., Tan, B.C.Y., Wei, K.K., 2005. Contributing knowledge to electronic Rodan, S., 2010. Structural holes and managerial performance: identifying the
knowledge repositories: an empirical investigation. MIS Quarterly 29, 113–143. underlying mechanisms. Social Networks 32 (1).
Kanter, R., 1985. Supporting innovation and venture development in established Rogers, E.M., Kincaid, D.L., 1981. Communication Networks: Toward a New Paradigm
companies. Journal of Business Venturing 1, 47–60. for Research. Free Press, New York.
Katz, D., 1964. The motivational basis of organisational behavior. Behavioural Sci- Schulz, M., 2003. Pathways of newness: exploring inflows of knowledge into sub-
ence 9, 131–146. units of multinational corporations. Organization Science 14, 440–459.
Kalish, Y., Robins, G.L., 2006. Psychological predispositions and network structure: Shah, P.P., 2000. Network destruction: the structural implications of downsizing.
the relationship between individual predispositions, structural holes and net- Academy of Management Journal 43 (1), 101–112.
work closure. Social Networks 28, 56–84. Simmel, G., 1950. The sociology of George Simmel. Free Press, Glencoe, IL.
Kaše, R., Paauwe, J., Zupan, N., 2009. HR practices, interpersonal relations, and Sparrowe, R.T., Liden, R.C., Wayne, S.J., Kraimer, M.L., 2001. Social networks and the
intrafirm knowledge transfer in knowledge-intensive firms: a social network performance of individuals and groups. Academy of Management Journal 44,
perspective. Human Resource Management 48, 615–639. 316–325.
Kelly, H.H., Thibaut, J.W., 1978. Interpersonal Relationships: A Theory of Interdepen- Stephenson, K., Krebs, V., 1993. A more accurate way to measure diversity. Personnel
dence. Wiley, New York. Journal 72, 66–76.
Klein, K.J., Lim, B., Saltz, J.L., Mayer, D.M., 2004. How do they get there? An examina- Sundgren, M., Dimenas, E., Gustafsson, J.-E., Selart, M., 2005. Drivers of organiza-
tion of the antecedents of centrality in team networks. Academy of Management tional creativity: a path model of creative climate in pharmaceutical R&D. R&D
Journal 47, 952–963. Management 35, 359–374.
Ko, D.G., Kirsch, L.J., King, W.R., 2005. Antecedents of knowledge transfer from con- Szulanski, G., 1996. Exploring internal stickiness: Impediments to the transfer of
sultants to clients in enterprise systems implementations. MIS Quarterly 29, best practice within the firm. Strategic Management Journal 17, 27–43.
59–85. Szulanski, G., 2003. Sticky Knowledge: Barriers to Knowing in the Firm. Sage, London.
Kowal, J., Fortier, M.S., 1999. Motivational determinants of flow: contributions from Teece, D., 1998. Capturing value from knowledge assets: the new economy mar-
self-determination theory. The Journal of Social Pscyhology 139 (3), 355. kets for know-how, and intangible assets. California Management Review 40,
Lin, H., 2007. Effects of extrinsic and intrinsic motivation on employee knowledge 55–79.
sharing intentions. Journal of Information Science 33, 135–149. Teece, D., Pisano, G., Shuen, A., 1997. Dynamic capabilities and strategic manage-
Mäkelä, K., Brewster, C., 2009. Interunit interaction contexts, interpersonal social ment. Strategic Management Journal 18, 509–533.
capital and the differing levels of knowledge sharing. Human Resource Manage- Teigland, R., Wasko, M., 2009. Knowledge transfer in MNCs: examining how intrinsic
ment 48, 591–613. motivations and knowledge sourcing impact individual centrality and perfor-
Marsden, P.V., 1990. Network data and measurement. Annual Review of Sociology mance. Journal of International Management 15, 15–31.
16, 435–463. Tichy, N.M., Tushman, M.L., Fombrun, C., 1979. Social network analysis for organi-
Marsden, P.V., 2002. Egocentric and sociocentric measures of network centrality. zations. Academy of Management Review 4, 507–519.
Social Networks 24, 407–422. Tsai, W., 2000. Social capital, strategic relatedness, and the formation of intra-
Marwell, G., Oliver, P., 1988. Social networks and collective action: a theory of the organizational linkages. Strategic Management Journal 21, 925–939.
critical mass ill. American Journal of Sociology 94 (3), 502–534. Tsai, W., 2002. Social structure of coopetition within a multiunit organization:
Mehra, A., Kilduff, M., Brass, D.J., 2001. The social networks of high and low coordination, competition, and intra-organizational knowledge sharing. Orga-
self-monitors: implications for workplace performance. Administrative Science nization Science 13, 179–190.
Quarterly 46, 121–146. Tsai, W., 2001. Knowledge transfer in intraorganizational networks: effects of
Mellahi, K., Wilkinson, A., 2008. A study of the association between downsizing and network position and absorptive capacity on business unit innovation and per-
innovation determinants. International Journal of Innovation Management 12 formance. Academy of Management Journal 44, 996–1004.
(4), 677–698. Tsai, W., Ghoshal, S., 1998. Social capital and value creation: the role of intra-firm
Moch, M.K., 1980. Job involvement, internal motivation, and employees integra- networks. Academy of Management Journal 41, 464–476.
tion into networks of work relationships. Organisational Behaviour and Human Tsang, E., 2002. Acquiring knowledge by foreign partners from international joint
Performance 25, 15–31. ventures in a transition economy: learning-by-doing and learning myopia.
Moreno, J.L., 1961. Role concept, a bridge between psychiatry and sociology. Amer- Strategic Management Journal 23, 835–854.
ican Journal of Psychiatry 118, 518–523. Tushman, M., 1977. Special boundary roles in the innovation process. Administrative
Mom, T.J.M., van den Bosch, F.A.J., Volberda, H.W., 2007. Investigating managers Science Quarterly 22, 587–605.
exploration and exploitation activities: the influence of top-down, bottom- Vallerand, R.J., Bissonnette, R., 1992. Intrinsic, extrinsic, and amotivational styles
up, and horizontal knowledge inflows. Journal of Management Studies 44, as predictors of behavior: a prospective study. Journal of Personality 60,
910–931. 599–620.
Moorman, C., Miner, A.S., 1998. Organizational improvisation and organizational Vallerand, R.J., 1997. Toward a hierarchical model of intrinsic and extrinsic motiva-
memory. Academy of Management Review 23, 698–723. tion. Advances in Experimental Social Psychology 29, 271–360.
Minbaeva, D., Pedersen, T., Bjorkman, I., Fey, C., Park, H., 2003. MNC knowledge Vallerand, R.J., 2000. Deci and Ryans self-determination theory: a view from the
transfer, subsidiary absorptive capacity and knowledge transfer. Journal of Inter- hierarchical model of intrinsic and extrinsic motivation. Psychological Inquiry
national Business Studies 34, 586–599. 11, 312–318.
634 R. Aalbers et al. / Research Policy 42 (2013) 624–634

Vallerand, R.J., Ratelle, C.F., 2002. Intrinsic and extrinsic motivation: a hierarchical Wasserman, S., Faust, K., 1994. Social Network Analysis: Methods and Applications.
model. In: Deci, E.L., Ryan, R.M. (Eds.), The motivation and self-determination Cambridge University Press, New York, NY.
of behavior: theoretical and applied issues. University of Rochester Press, Weir, D., Hutchings, K., 2005. Cultural embeddedness and contextual constraints:
Rochester, NY. knowledge sharing in Chinese and Arab cultures. Knowledge and Process Man-
van Driel, H., Dolfsma, W., 2009. Path dependence, initial conditions, and routines agement 12, 89–98.
in organizations: the Toyota production system re-examined. Journal of Orga- Wellman, B., Berkowitz, S.D., 1988. Social Structures: A Network Approach.
nizational Change Management 22 (1), 49–72. Cambridge University Press, Cambridge, MA.
Vansteenkiste, M., Simons, J., Lens, W., Sheldon, K.M., Deci, E.L., 2004. Motivating Whelan, E., Parise, J., de Valk, J., Aalbers, R., 2011. How scouts and connec-
learning, performance, and persistence: the synergistic effects of intrinsic goal tors can deliver open innovation. MIT Sloan Management Review 53 (1),
contents and autonomy supportive contexts. Journal of Personality and Social 37–44.
Psychology 87, 246–260. Woodman, R.W., Sawyer, J.E., Griffin, R.W., 1993. Toward a theory of organizational
Wasko, M.M., Faraj, S., 2000. It is what one does: why people participate and help creativity. Academy of Management Review 182, 293–321.
others in electronic communities of practice. Journal of Strategic Information Yang, J.T., 2004. Job-related knowledge sharing: comparative case studies. Journal
Systems 9, 155–173. of Knowledge Management 8 (3), 118–126.
Wasko, M.M., Faraj, S., 2005. Why should I share? Examining social capital and Zollo, M., Winter, S.G., 2002. Deliberate learning and the evolution of dynamic capa-
knowledge contribution in electronic networks of practice. MIS Quarterly 29, bilities. Organizational Science 13 (3), 339–351.
35–57.