Testing impact of knowledge characteristics and relationship ties on project performance

Jina Kang

Abstract Purpose The aim of this paper is to empirically test and nd the correlation between knowledge characteristics and relationship ties on project performance. Design/methodology/approach Data were collected via personal interviews based on a structured survey of project managers in a knowledge-intensive rm. The data were analyzed using a multiple regression model. Jina Kang is Assistant Professor, Graduate Program of Technology and Management and Department of Industrial Engineering, Seoul National University, Seoul, South Korea. Findings The results show that project performance was positively related to the frequency and closeness of source, and difculty of the knowledge element described in the survey. Interestingly, against the prevalent view, the source and the level of tacitness of the knowledge element were not found to be signicant. Research limitations/implications Data are limited to specic corporate setting and variables are not exhaustive, despite the fact that this study includes the most theoretically interesting variables. One implication is that strong ties to a knowledge element source are important for project success, but distinctions between internal and external sources, and degree of codication, are not. Practical implications This study also implies that, in this day and age of knowledge management and chief knowledge ofcers, most of the important knowledge may already have been codied inside the rm, thus alleviating the past focus on transfer of tacit knowledge. This fact also implies that rms that are left behind in terms of managing their knowledge inside the rm stand to lose a lot more since the other rms are managing their own knowledge base better. Originality/value This paper incorporates and empirically tests most of the variables considered important to theoreticians and practitioners in the realm of knowledge management and network theory. The latter theory is important in the eld of knowledge management because it offers the link and medium in which knowledge travels and transfers. Keywords Knowledge management, Knowledge sharing, Networking, Tacit knowledge Paper type Research paper

Introduction
In recent years, effective management of knowledge has become a critical task in building competitive advantage (Teece, 1996; Eisenhardt and Santos, 2001). This steady shift in interest from physical resources to an abstract resource called knowledge reects the signicant disparities in performance across teams and divisions within a single rm when the necessary knowledge is not shared effectively within the rm. Firms that capture useful knowledge about their practices, and are able to share it across the organization, can sharpen their competitive edge. For example BP, a $262 billion oil giant, cut down drilling costs by $47 million per oil well in 1998 by identifying and sharing seemingly insignicant innovations that workers practiced in different parts of the world (Prusak, 1999). The knowledge-based view draws its roots from the resource-based view, organizational learning, and dynamic capabilities. These literatures explain what knowledge is and why knowledge is critical to competitive advantage. Social network theory gives us insight into how knowledge moves between individuals of the rm, because knowledge by itself does

This research beneted from very helpful comments by Phillip Bonacich, Martin Dierker, Richard Goodman, Marvin Lieberman, Olav Sorenson, and Burt S. Swanson. The author thanks the Anderson School at UCLA for nancial support, and Yujin Han and Hyojung Lim of SNU for their research assistance.

PAGE 126

JOURNAL OF KNOWLEDGE MANAGEMENT

VOL. 11 NO. 3 2007, pp. 126-144, Q Emerald Group Publishing Limited, ISSN 1367-3270

DOI 10.1108/13673270710752153

not move. Other research draws on transaction cost economics to explore the relationship between a rms organization and management of knowledge. However, as Argyres and Silverman (2004) note, and exemplify, most research focus on knowledge in an inter-rm setting, almost to the exclusion of intra-rm setting. More infrequent and rare is research encompassing both inter- and intra-rm settings to ensure a balanced look at knowledge and how it behaves in these different environments. The purpose of this paper is to precisely address this hole in the research eld and to nd how knowledge behaves in both intra- and inter-rm settings. The resource-based view of the rm explains rm performance differences that the structure-conduct-performance paradigm of industrial organization (IO) (Bain, 1959; Porter, 1980) does not consider. Wernerfelt (1984), Dierickx and Cool (1989), and Prahalad and Hamel (1990) built a resource-based theory based on the idea of competitive advantage rooted inside a rm rather than from external attributes of its industry. Some sources and nature of competitive advantage of the rm have since been suggested, such as: valuable, rare, inimitable, and non-substitutable resources (Wernerfelt, 1984; Barney, 1991); the managements abilities to marshal these assets (Grant, 1991); and related sets of operational routines and technological skills (Prahalad and Hamel, 1990; Stalk et al., 1992). Added to the idea of competitive advantage of the rm is the idea that in an increasingly fast-moving environment, very few competitive advantage can be sustained over time unless the rm has organizational and managerial processes termed dynamic capabilities to integrate, build, and recongure internal and external competencies (Teece et al., 1997; Eisenhardt and Brown, 1998). Thus given the increasingly dynamic nature of the environment itself, resources static in nature such as xed assets or outdated skills can no longer be deemed a sustainable competitive advantage of the rm. Based on this understanding of organizations and environments, Grant (1996) proposed that tacit knowledge is the source of sustained competitive advantage. He argues that the source of competitive advantage in dynamic environments is not knowledge that is codied and proprietary to the organization, because the value of such knowledge erodes quickly due to obsolescence and imitation. Rather, sustained competitive advantage is determined by non-proprietary knowledge in the form of tacit individual knowledge. According to Grant (1996), tacit knowledge can form the basis of competitive advantage because it can be both unique and relatively immobile. Yet, because that knowledge is possessed by individuals and not the organization, a critical element of sustained competitive advantage is the ability to integrate the specialized and tacit knowledge of individuals. The concept and types of knowledge, originally borrowed from sociology, has since been expanded to answer the question of why rms exist. Conner and Prahalad (1996) identify knowledge carried and shared by individuals within the rm to be both the raison detre for that rm against alternative forms of organization such as markets, and also to be the source of such rms performance difference with other rms. Kogut and Zander (1996) also expand on this belief that the rm is a better facilitator of knowledge sharing than a market, hence its predominance in the world. Despite the preponderance of the tacitness of knowledge as a sustained competitive advantage of the rm (Nelson and Winter, 1982), recent views from mostly European schools of thought suggest that codied knowledge is a more solid source of rm competitiveness (Sanchez and Heene, 1997). The characteristic of difculty in transmitting tacit knowledge

This study, by focusing on knowledge in its most basic form, will unravel some of the organizational-level issues involved with knowledge management.

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 127

has led scholars to assert that tacit knowledge in particular is a source of competitive advantage because it is more difcult for other rms to copy or replicate (Teece and Pisano, 1994; Teece et al., 1997). But on the other hand, the same nature of tacit knowledge that makes it difcult for other rms to copy also makes it difcult to connect internally. Thus Sanchez and Heene (1997) posit that it is not tacit knowledge, easily lost through employee turnover and job switching, but it is the explicit knowledge that has been codied, which builds the foundation for rm competitiveness. Sanchez and Heene (1997) argue that Polyanis (1962) use of the term know in the famous example of the skater who knows more than she can tell about how she skates. They suggest that whatever the skater cannot tell (albeit with some mental effort), she really does not know, at least not within the concept of knowledge proposed by Sanchez, Heene and Thomas (1996). Sanchez et al. (1996) seem to imply that tacit knowledge cannot be relied on to improve rm performance, and only explicit knowledge can be used to compete with others. To them, knowledge is the set of beliefs held by an individual about causal relationship among phenomena (Sanchez et al., 1996, p. 9). They argue that strategically relevant knowledge is never certain, but rather only exists in the form of beliefs, in the sense that knowledge is not absolute or deterministic, but consists of probabilistic assessments of causal relationships between phenomena. They agree with Polyani in the aspect that knowledge originates with and exists within individual humans, but expand their denition of knowledge by conceding that organizations may also have knowledge that may exist in various forms understood by more than one individual within an organization (Sanchez et al., 1996, p. 5). This sharing of knowledge by more than one individual form the basis of that organizations competence, which Sanchez et al. (1996, p. 8) dene as:
[. . .] an ability to sustain the coordinated deployment of assets and capabilities in a way that promises to help a rm to achieve its goals.

Thus the degree of codiability is yet to be empirically tested in its contribution to rm competitiveness. Another dimension of knowledge, its difculty, is another salient way of categorizing knowledge. Difculty is considered one of the aspects having the most inuence in the transfer of knowledge (Rogers, 1962; Winter, 1987). Knowledge by itself in isolation cannot move. It requires a medium, and the medium is usually individuals or products created by individuals, such as books or software. This requirement of a medium necessitates individual interaction, meaning ties between people or network of ties. This is where the organizational learning literature comes in to explain the knowledge-based view. Learning can be dened as the process by which new information or knowledge is incorporated into the behavior of agents, changing their patterns of behavior. Brown and Duguid (1991) proposed a unied view of working, learning, and innovation which links individual and organizational levels of knowledge. The authors start by pointing out that codication of work procedures can be quite different from actual working practices, even contradictory. They argue that learning theory should be distanced from codied, transferable, and objective notions of knowledge, and focus instead on knowledge in context. In their view, meaningful knowledge is deeply related to daily work, and the acquisition of new knowledge (i.e. learning) is socially constructed from working practices. This social construction of knowledge occurs within informal communities, where knowledge is freely shared through collaborative mechanisms such as communications and joint work. Nelson and Winter (1982) were among the rst to integrate organizational knowledge and routines with the notion of dynamic competitive environments. In their approach to evolutionary economics, the rm is understood to be a repository of knowledge, which is represented by routines that guide organizational action. The authors see individuals as responding to information difculty and uncertainty through their own skills and routine organizational activity, in line with the behavioral tradition (Simon, 1965; Cyert and March, 1963). Cohen and Levinthal (1990) related organizational learning and innovation to the evolving knowledge base of the rm. The authors dene absorptive capacity as the ability to

PAGE 128 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007

 Project performance is strongly associated with high frequency and high closeness of the relationship tie, and the difculty of the knowledge element.

recognize the value of external information, assimilate it, and apply it to commercial ends. According to the authors, absorptive capacity is largely a function of the level of the rms prior knowledge (which emphasizes the cumulative nature of knowledge), and is history or path-dependent (which emphasizes the importance of earlier decisions). Important determinants of absorptive capacity are the internal channels of communication, the distribution of knowledge in the environment and in the rm, and the pattern of R&D investment decisions. To describe the internal channels of communication that Cohen and Levinthal (1990) identify as an important determinant of absorptive capacity, theories in network literature are helpful. Initially Granovetter (1973) advanced the idea that weak ties infrequent and distant relationships facilitate searching for new information because their contacts are less likely to provide redundant knowledge. He posited and found evidence to support the proposition that weak ties serve as a source of new information, or explicit knowledge. However, he later revisited the idea of tie strength and noted that strong ties have greater motivation to be of assistance and typically more easily available than weak ties (Granovetter, 1982, p. 113). Krackhardt (1992) extended this work to suggest that the benet of strong ties is that they provide a base of trust that is crucial when an actor or individual confronts uncertainty. Thus strong ties facilitate innovation by providing exchanges that reduce suspicion. The strength of ties is usually dened by a few characteristics. Among the prominent are: frequent interaction (frequency) and intimacy (closeness) (Granovetter, 1982) between the parties to the relationship (Krackhardt, 1992). Strong ties are thought to provide less diverse or novel information because individuals are much more likely to form strong ties with socially similar individuals, who tend to possess the same information and to hold similar opinions (Granovetter, 1973; Rogers, 1962). However, they do provide other benets that may facilitate adaptation. First, strong ties are more likely to promote in-depth two-way communication and to facilitate the exchange of detailed information (Granovetter, 1982; Krackhardt, 1992; Uzzi, 1996). Thus although strong ties may not maximize the awareness of environmental changes and potential adaptive responses, they are likely to be more valuable than weak ties in helping to decipher the implications of external threats and to evaluate potential responses to these threats. Further, the trust and mutual identication that are likely to exist when ties are strong make it more likely that both information will be shared and actually acted on. This study is based on structured survey interviews with top managers of project teams of a rm. Project teams have long been the focus of research, especially in product development (Myers and Marquis, 1969; Rothwell, 1972; Imai et al., 1985). Product development is one practical case of applying knowledge to create new knowledge or innovation. In many rms, project team members are the people who actually do the work of product development (Brown and Eisenhardt, 1995). These are the people who transform vague ideas, concepts, and product specications into the design of new products. Thus it is not surprising that project team is central to many models of product development (Brown and Eisenhardt, 1995). The unit of analysis in this study is a knowledge element, dened as the smallest unit of discrete and independent piece of knowledge necessary to carry out an operation. The following hypotheses are derived from the theoretical discussion and tested:

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 129

H1.

The closer the source of knowledge is to the individual of the project team, the higher the knowledge elements importance to its project performance (source of knowledge). The higher the tacitness component of the knowledge element, the higher the knowledge elements importance to its project performance (tacitness of knowledge). The higher the difculty component of the knowledge element, the higher the knowledge elements importance to its project performance (difculty of knowledge). The higher the frequency of contact to the source of knowledge during the course of the project, the higher the knowledge elements importance to project performance (frequency of ties). The closer the contact of knowledge source is to the individual at the beginning of the project, the higher the knowledge importance to project performance (closeness of ties).

H2.

H3.

H4.

H5.

This study focuses on how knowledge is integrated from different sources internal and external to generate new knowledge and thus contribute to project performance of a rm. This is unique from other studies in its integrative nature. Most research in knowledge theory research centers on one aspect of knowledge sourcing, knowledge integration, internal knowledge transfer, or external knowledge transfer (Eisenhardt and Santos, 2001) or either inter-rm setting or limitedly, intra-rm setting (Argyres and Silverman, 2004). Studies on knowledge sourcing suggest that external linkages are important for a variety of innovation-related outcomes. These external linkages include incentives that motivate scientists to stay connected with the larger scientic community (Henderson, 1994), formal network relationships (Powell et al., 1996; Brown and Eisenhardt, 1997; McEvily and Zaheer, 1999), exploratory products (Brown and Eisenhardt, 1997, Jett, 1999), gatekeepers (Allen, 1977; Katz and Tushman, 1981), and informal networks (Henderson and Cockburn, 1994; Liebeskind et al., 1996; Tripsas, 1997; Rosenkopf and Nerkar, 2001; Cummings, 2004). Few studies suggest that similar knowledge sourcing processes occur within rms and can also lead to more innovation (Henderson and Cockburn, 1996; Hansen, 1998). Yet this research stream leaves unexamined several fundamental issues pertinent to strategy studies. Only a few studies examine rm performance, and those that do rely on a variety of close proxies, including survival (Tripsas, 1997), growth (Powell et al., 1996), market segment dominance (Brown and Eisenhardt, 1998), and prot (Bierly and Chakravarti, 1996). As a result, these need no demonstration of any specic conception of performance (Eisenhardt and Santos, 2001). Further unlike this study, these studies do not examine micro-processes within the rm failing to give us a clear picture of how knowledge actually creates value within the rm. For this we have to look at another area of empirical research: knowledge integration. Studies with a focus on knowledge integration emphasize the micro-processes of interaction, mostly within organizations, that enable the integration of knowledge. This stream of research is especially relevant to knowledge theory because of the assertion that knowledge integration (especially integration of the tacit knowledge that is held by individuals) is a primary source of superiority of rms over markets and a major way in which competitive advantage is achieved (e.g., Grant, 1996). However, most studies in this eld are ethnographic (e.g., Hargadon and Sutton, 1997; Bechky, 1999; Helper et al., 1999) or experimental (Okhuysen and Eisenhardt, 2002) in nature, thus failing to address rm performance, unlike this study. Most studies on internal knowledge transfer tend to focus on the knowledge characteristics that affect the efcacy of internal knowledge transfer. These characteristics include tacitness (Zander and Kogut, 1995; Lord and Ranft, 1998), causal ambiguity (Szulanski, 1996), and complexity (Hansen, 1998), which impede knowledge transfer, and strategic value (Brown and Eisenhardt, 1998; Gupta and Govindarajan, 2000), which enhances knowledge transfer. Some of these internal focused studies also included relationships. These studies also

PAGE 130 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007

indicate that the relationship between the sender and recipient is crucial for knowledge transfer. When the sender and recipient have difculty establishing interpersonal interactions (Szulanski, 1996) such as when they are distant, knowledge transfer is impaired. In contrast, when integrative mechanisms such as teams, informal social networks, norms for collaboration, and formal meetings exist (e.g., Hargadon, 1998; Gupta and Govindarajan, 2000; Eisenhardt and Galunic, 2000), knowledge transfer is facilitated. Some studies looked at research units (Birkinshaw et al., 2002; Hansen et al., 2005), informal relationships (Levin and Cross, 2004) and globally dispersed R&D organizations (Kim et al., 2003) with some social science approach of looking at researchers within the rm (Reagans and Zuckerman, 2001). In summary, these studies are very useful for understanding internal knowledge transfer. Lastly, studies on external knowledge transfer indicate that knowledge transfer is affected by knowledge characteristics and by the relationship between the sender and recipient. As such these studies replicate the studies of internal knowledge transfer in terms of its method, variable contented, and results except in some alliance relationships (Mowery et al., 1996; Powell et al., 1996). The studies, unlike this one, do not try to answer whether external knowledge transfer is either easier or qualitatively different from internal knowledge transfer. Some consider this a theoretically critical question because a primary assumption of knowledge theory is that knowledge transfer is facilitated within rms as compared with markets (Kogut and Zander, 1996; Grant, 1996). While some notable effort has undergone to link knowledge management and R&D management for example (Park and Kim, 2005), much is needed at intra- and inter-rm settings like this article does. This study of knowledge management is uniquely positioned at the knowledge element level to nd its impact on project performance. Knowledge element is dened as a discrete and independent piece of knowledge. During the interview, respondents are given the denition and examples of the knowledge element. The examples of a knowledge element are: information (previously available ndings, such as data sets, literature, and published reports); analytical methods; any insights; any measurement methods; any project management knowledge; client knowledge; any client management knowledge; and others. Most studies use a chunk of knowledge piece such as an innovation dened simply as new knowledge or expertise (Tsai, 2001, p. 1000). Most studies tend to use a conceptual concept of knowledge that is not quantied or even dened (e.g., Hansen, 1999; Kogut and Zander, 1993). To explore the extent of knowledge importance to a project, the author considers the following three attributes of knowledge: 1. the characteristics of knowledge (tacitness and difculty); 2. the source of the knowledge (internal or external to the rm); and 3. the strength of the connection to the source (strength of tie, or frequency or closeness of ties). The knowledge elements are a unit of analysis and may have different characteristics such as difculty or degree of tacitness, which are some of dependent variables. This study, by focusing on knowledge in its most basic discrete form, will unravel some of the organizational-level issues involved with knowledge management.

Tacitness and source of the knowledge element were not found to be signicant.

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 131

Method
The data for this study were collected by personal interviews based on a structured survey. Project managers of a knowledge-intensive rm were interviewed to answer questions regarding knowledge characteristics, ties, and project performance. Prior to the interviews, pre-tests were conducted using a small but similar sample. This pre-test phase allowed the survey to be revised and interview technique to be rened. The survey embedded several advances recommended by previous interview researchers (e.g., Cronbach, 1984; Diesing, 1971; Sundberg, 1977; Nunnally, 1978; Corcoran and Fischer, 1987). Data collection and research site This study is based on structured survey interviews with top managers of project teams of a rm. The project managers were selected as respondents because they are the ones who are in the thick of things (Hansen, 1999). The author tested the predictions in a large, multidepartmental and multinational consulting company (hereafter called the rm or the company). The company, which has annual billing of more than $5 million, is involved in developing, planning, and selling a range of consulting analyses. It has been protable for a number of years and has continued to grow. The company is structured into a number of fairly autonomous departments that are responsible for business. These departments are organized according to product-market segments or industries, such as medicine, education, and heavy industries. Within each department are project teams, formed for each new project by a designated project manager. The project manager has the option of selecting team members based on the needs of the project which may span over years of data collection and multinational collaboration. Table I gives a brief description of the projects involved. Model A simple ordinary linear regression (OLS) was used to analyze the data collected from the personal interviews. Often it is the simplest model that offers most insight and understanding of a phenomenon. The model is as follows. The contribution of a knowledge element i of project j to the project performance is assumed to be given by: R ij b0 b1 Locselfij b2 Freqij b3 Closeij b4 Tacitij b5 Diffij d1 Durationj d2 Sizej 1ij for all i and j; where E 1ij 0 and cov1ij ,1kl 0 whenever i k or j l. The LocSelf variable refers to the source of knowledge. The respondents were asked to identify the source of knowledge amongst the following alternatives: 1 from outside of the rm, 2 distantly located department, 3 from unrelated department, 4 from related department, 5 from the team, 6 from a team member, 7 personal knowledge. Thus the higher the beta coefcient, the more internal is the source of the knowledge element. In addition, regressions are done with the LocSelf variable split into three parts: personal knowledge (a dummy variable called LocDumSelf); knowledge from outside the rm (a dummy variable named LocDumExt); and inside the rm (a variable called LocInFirm of four-interval scale from 2 distantly located department to 6 from a team member). The Freq variable refers to the frequency of contact with the source of knowledge. The respondents are asked to choose the frequency of knowledge among the following alternatives: 1 less than once a month, 2 twice a month, 3 thrice a month, 4 once a week, 5 twice a week, 6 thrice a week, 7 essentially every working day. Thus the higher the beta coefcient, the higher the frequency of contact with the source of the knowledge element. The Close variable refers to the closeness of contact with the source of knowledge. The respondents are asked to indicate the closeness of knowledge from a continuous scale

PAGE 132 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007

Table I Models with third transformation (restricted set of four top elements)

Variable 0.44** 0.07 0.05 0.05 0.06** 0.06*** 0.01 0.05 0.14 0.19 0.05 0.05 0.06** 0.06*** 0.98 0.44**

Model 19 Standard error n n n 0.94

Model 20 Standard error

Model 21 Standard error

Model 22 Standard error 0.51*

n 1.00 0.03 0.04 0.12 0.19

Model 23 Standard error 0.42** 0.05 0.05 0.06** 0.06***

Model 24 Standard n error 0.93 0.50*

1.09 20.03 0.02 0.05 0.13 0.19

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 133

j
0.18 20.45 0.11 0.09

1.03 20.03 0.02 0.05 0.13 0.20 20.00 0.02 0.23 0.45 0.11

0.52* 0.07 0.05 0.05 0.06 0.06*** 0.01 0.06

0.01 0.05 0.15 0.19 20.00 0.02 0.20 20.47 0.10

0.05 0.05 0.06** 0.06*** 0.01 0.06 0.24 0.46

0.03 0.04 0.12 0.19 20.00 0.02

0.05 0.05 0.06** 0.06*** 0.01 0.06

Constant LocSelf Freq Close Tacit Diff Duration Size LocDumSelf LocDumExt LocInsFirm Adj R-square

0.03 0.11

0.07

0.03 0.09

0.07

Notes: * p , :010; ** p , 0:05; *** p , 0:01; values are unstandardized regression coefcients

where 1 not at all close, 4 moderately close, 7 very close. Thus the higher the beta coefcient, the closer was the source of the knowledge element to the respondent. The Tacit variable refers to the tacitness of the knowledge element. The respondents are asked to indicate the tacitness from a continuous scale where 1 mainly reports manuals documents self-explanatory software etc., 4 half know-how and half reports or documents, 7 mainly personal practical know-how, trick-of-trade. Thus the higher the beta coefcient, the more tacit the knowledge element. The Diff variable refers to the difcult of the knowledge element. The respondents are asked to indicate the difculty using a continuous scale where 1 not at all difcult or difcult, 4 moderately difcult or difcult, 7 very difcult or difcult. Thus the higher the beta coefcient, the more difcult the knowledge element. Dependent variable. The dependent variable Rij captures the importance of the knowledge elements contribution to its projects performance. After identifying knowledge elements for a project, respondents were then asked to rank-order the knowledge elements in their importance to the projects success. The rankings are then converted to a fraction for the following reasons. The number of knowledge element varies from project to project. This may distort the relative importance of knowledge elements in projects with less number of knowledge elements to those with more knowledge elements. This transformation corresponds to [(number of knowledge element in the project 2 ranking 1)/number of knowledge element in the project] or where kth ranked element of n elements is n 2 k 1=n. This number, which is a fraction greater than 0 and equal to or less than 1, is referred to as the Rank variable. This also ensures that the ordering of the dependent variable is compatible with that of the other variables in the model. For example, a knowledge element most critical in a project would always be assigned the value 1 and the least important knowledge element would always be assigned the value of (1/number of knowledge element in the project). In addition to this original transformation, three other transformations are conducted to ensure robustness of the dependent variable. The rst transformation is where kth of n elements is 2n2k =2n 2 1, whose characteristics among others, is that a ranking would carry twice as more weight than the previous ranking. The second transformation is a clustering of knowledge elements into three parts: the top third, the middle third, and the bottom third of knowledge elements in a project. The third transformation included only the top four knowledge elements. Control variables Next, ordinary linear regression was performed with the control variables included. There are two control variables and they are called Duration and Size. The Duration control refers to the duration of the project in months. The Size control refers to the number of people involved in the team. Summary statistics Table II presents the summary statistics. The 228 knowledge element observations were collected from 27 projects of the rm. The average number of knowledge elements from a project is 9.7, distributed between three elements and 15 elements. Results In the regression model including the control variables, Freq, Close, and Diff variables are shown to be signicant, with beta coefcients of 0.037, 0.022, and 0.061 respectively. Adjusted R-square is 0.24 (see Table III). Therefore the hypotheses regarding frequency, closeness, and difculty of knowledge elements are supported by the data, but the hypotheses regarding the source of the knowledge element and the tacitness of knowledge element cannot be supported.

PAGE 134 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007

Table II Description and correlations of key variables

Variable 1. 2. 3. 4. 5. 6. 7. 8. Rank LocSelf Freq Close Tacit Diff Duration Size Mean 0.57 5.62 3.33 4.19 4.75 3.88 33.62 5.45 SD 0.29 1.57 1.89 2.02 1.90 1.65 18.64 2.18 Max. 1.00 7 7 7 7 7 60 10 Min. 0.07 1 1 1 1 1 4 2 1 2 3 4 5 6 7

20.01 0.36 0.15 20.09 0.41 20.01 20.02

0.11 0.09 0.13 0.07 0.06 0.06

0.12 20.10 0.29 20.08 20.01

0.08 20.06 20.21 20.05

0.01 0.03 0.11

0.08 0.04

0.42

Increasing the frequency of contact is associated with increased importance of the knowledge gained from that contact. The regression results also imply that difculty of knowledge is associated with the importance of that knowledge to project performance. This study shows that project performance is strongly associated with high frequency, high closeness, and high difculty of the knowledge element. However, the actual source of knowledge and the tacitness of knowledge element were found to be irrelevant in explaining project performance. The results show strong support for strong relationships and project performance. Along with the original model, the various location variables are tested in two different models, the rst model with LocDumSelf and LocDumExt, and the second model with LocInFirm. Results show a similar pattern of signicance and coefcients to the original model. The Freq, Close, and Diff variables stay signicant and the Tacit variable along with LocDumSelf, LocDumExt, and LocInFirm stay insignicant like the LocSelf variable in the original model. Even the R-square stays in the same range. The similarity of pattern in signicance and coefcients among the models seems to ensure the robustness of the original model. However, some interesting patterns emerge when different forms of transformations are used for the dependent variable. The additional rst and second transformations led to a similar pattern of signicance amongst variables but lower R-square than the original model. However, the third transformation led to an interesting result. In the model using the third transformation of the Restricted Sample Model (the transformation which include only the top four knowledge elements), the Tacit variable shows signicance, whereas the Close and Freq variables do not show signicance. The Diff variable stays signicant in this model (see Tables I, IV and V).

Discussion and conclusion

In this article, the author reviewed various theories of knowledge and social network theory and identied several characteristics of knowledge and relationship ties considered important in building competitive advantage of a rm. Each of the characteristics was then operationalized and surveyed during face-to-face interviews based on structured interviews. The results showed that project performance is strongly associated with high frequency and high closeness of the relationship tie, and the difculty of knowledge element; however, tacitness and source of the knowledge element were not found to be signicant. These lead to insightful implications, shifting the focus to relationship ties and the type of interactions from the type of knowledge, typied by tacitness of knowledge. This study suggests that tacitness of knowledge does not have any relationship to project performance previously theorized by many in the eld, but rather uphold the views of Sanchez and Heene (1997). Tacitness has been a strong idea in the past literature. Many in the eld concede that tacitness is difcult to measure, difcult to describe, and therefore difcult to transfer, even inside an organization. Thus the importance of tacitness in competitive advantage, according to the mainstream view of knowledge, has been tantamount to competitive advantage itself.

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 135

PAGE 136 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007

j
n 0.18 0.04 0.02 20.01 0.06 0.01** 0.01* 0.01* 0.01 0.07* 0.18 Model 2 Standard error n n Model 3 Standard error Model 4 Standard error 0.09 n 0.16 0.04 0.02 0.01 0.06 Model 5 Standard error 0.07 0.00** 0.00* 0.00 0.01** n 0.17 0.04 0.02 0.01 0.06 0.00 0.00 0.04 0.07 0.24 Model 6 Standard error 0.09 0.00** 0.00* 0.00 0.01** 0.00 0.00 0.08** 0.01 0.01** 0.01** 0.01 0.01** 0.23 0.01 0.04 0.02 20.01 0.06 0.00 0.00 0.03 0.02 0.24 0.24 0.09* 0.01 0.00** 0.00* 0.00 0.01** 0.00 0.00 0.04 0.02 20.01 0.06 0.00 20.00 20.03 0.03 0.23 0.00** 0.00* 0.00 0.01** 0.00 0.00 0.04 0.07 0.01 0.24 0.01 0.01 0.24 0.01

Table III OLS regression models of the performance for knowledge characteristics

Variable

Model 1 Standard error

Constant LocSelf Freq Close Tacit Diff Duration Size LocDumSelf LocDumExt LocInsFirm Adj R-square

0.24 0.01 0.04 0.02 0.01 0.06

Notes: * p , 0:10; ** p , 0:01; values are unstandardized regression coefcients

Table IV Models with rst transformation (kth of n elements is 2n2k =2n 21)

Variable 0.05 0.00 0.00** 0.00** 0.00* 0.00*** 20.02 0.01 0.01 20.01 0.03 0.01** 0.01* 0.01 0.01*** 20.01 20.00 0.01 0.01 20.01 0.03 0.00 20.00 20.01 20.03 0.13 0.13 0.13 0.02 0.05 0.04

Model 7 Standard error n n n 20.01

Model 8 Standard error

Model 9 Standard error

Model 10 Standard error 0.05

Model 11 Standard error 20.04 0.01 0.01 20.01 0.03 0.05 0.01** 0.01* 0.01 0.01***

Model 12 Standard n error 20.03 0.01 0.01 20.01 0.03 0.00 20.00 0.05 0.01** 0.01* 0.01 0.01*** 0.00 0.01

20.02 20.00 0.01 0.01 20.01 0.03

0.06 0.01 0.01** 0.01** 0.01 0.01*** 0.00 0.01

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 137

0.01 0.01 20.01 0.03 0.00 20.00 20.01 20.03 0.12

0.01* 0.01** 0.01 0.01*** 0.00 0.01 0.02 0.04

Constant LocSelf Freq Close Tacit Diff Duration Size LocDumSelf LocDumExt LocInsFirm Adj R-square

0.01 0.13

0.01

0.01 0.13

0.01

Notes: * p , :010; ** p , 0:05; *** p , 0:01; values are unstandardized regression coefcients

PAGE 138 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007

j
n 0.85 0.08 0.07 20.03 0.18 0.03** 0.02** 0.03 0.03** 0.21** 0.78 Model 14 Standard error n n Model 15 Standard error Model 16 Standard error 0.25** n Model 17 Standard error 0.91 0.08 0.06 20.03 0.17 0.21** 0.03** 0.02* 0.03 0.03** Model 18 Standard n error 0.86 0.08 0.06 20.04 0.17 0.00 0.01 0.11 0.21 0.21 0.25** 0.03** 0.03* 0.03 0.03** 0.00 0.02 0.25** 0.03 0.03** 0.02** 0.03 0.03** 1.02 20.04 0.08 0.07 20.03 0.17 0.00 0.01 0.02 0.28 0.21 0.20 0.27** 0.03 0.03** 0.03** 0.03 0.03** 0.00 0.02 0.08 0.07 20.03 0.18 0.00 0.01 0.01 0.29 0.03** 0.03** 0.03 0.03** 0.00 0.02 0.11 0.22 20.00 0.21 0.03 20.00 0.20 0.03

Table V Models with second transformation (three clusters)

Variable

Model 13 Standard error

Constant LocSelf Freq Close Tacit Diff Duration Size LocDumSelf LocDumExt LocInsFirm Adj R-square

1.08 20.04 0.08 0.06 20.03 0.17

Notes: * p , 0:05; ** p , 0:01; values are unstandardized regression coefcients

However, the results of tacitness not showing signicant contribution to project performance is consistent with what Sanchez and Heene (1997) have been theorizing. Rather, the results of this study reinforce the idea that the interaction pattern does matter. This reect the recently highlighted spotlight on networking by rms, especially among rms that traditionally have been focusing on science and technology and less if any at all on human interactions. Indeed, engineering rms and even schools are now redirecting their energy on management issues to help understand organizational issues as well as interpersonal relationships. In this study, project performance is strongly associated with high frequency and high closeness of the relationship tie. It is interesting to note that the variable of closeness measures psychological distance rather than physical distance or work distance, as one could presume. Even before this new interest in management issues were highlighted, rms have been keen on sending their science workforce to academic and professional conferences to update their knowledge base and even publish their own ndings in academic or industry journals. This study shows that these efforts are and have been worthwhile. Tacitness may play a relatively smaller role in this context, but the conclusion may not be representative. In these days of knowledge management and chief knowledge ofcers, important knowledge may already have been codied inside the rm, thus alleviating the importance for tacit knowledge transfer. It may also well be that for a well-established rm, the crucial stage is in locating the knowledge. Once the knowledge is located, a well-established rm with enough resources, nancial or human, can easily procure the knowledge, either by learning or purchasing it. It is also interesting to note that there may be industries where tacit knowledge is less important than other industries. For example, a consulting rm may be different from a biotech rm in terms of the importance of tacitness. Consulting rms have been much more oriented towards explicit codication due to the extreme mobility of manpower. These are some vast areas left to be explored by future research. The reader should also be mindful of the fact that this particular rm where the study takes place is a consulting rm whose primary product is reports. This rm by nature, tradition, and evaluation produces a large number of internal and external reports for internal sharing as well as for clients. While the rm does not have a formal computerized knowledge management system, it has a long history of informal and formal sharing work, data, and methods through an efcient library system. In addition, employees of this rm have a relatively long tenure at the rm, meaning that work of colleagues is well and widely known. Many of both types of reports will nd their way to professional journals, where they will be ultimately shared by the entire community. All of this suggests that tacit knowledge plays a less of a role in this rm than in other rms even in the consulting industry. This implication may sound counter-intuitive initially in view of the mainstream knowledge theory, but the research on search costs may shed some light on this matter. Search costs the time and money spent locating the best product at the best price are a familiar and often unwelcome aspect of everyday life. Think of the teeth-gritting experience of driving from dealership to dealership in search of the perfect car. Even with the advent of e-commerce with vast amounts of information readily available, which seemed to predict slashed prices and razor-thin margins, recent studies suggest that the internets impact on pricing has been less dramatic especially online than some may have expected (Kwak, 2001). This study suggests several implications for management practice. Foremost, it is important to identify and locate knowledge. Because the effects of knowledge ow on project performance appear to be so consistently positive, CEOs and managers must be sure that important rm knowledge is stored and shared in the rm (Kogut and Zander, 1995). After the mid-1990s, many rms implemented knowledge management systems trying to codify their rm-specic and industry-general knowledge. In their effort to codify their knowledge, most rms fall into the trap of capturing only codiable knowledge, which may

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 139

 Instead of an intricate and user-friendly knowledge management system, rms should also invest in more informal employee gatherings such as lunches or outdoor activities.

be lower in the priority or importance. What they are doing may only exacerbate their competitiveness in the face of imitation and copying by their competitors. By focusing on a formal knowledge management systems, these rms may be missing out on important knowledge accrued by rst-hand experience. Instead of an intricate and user-friendly knowledge management system, rms should also invest in more informal employee gathering such as lunches or outdoor activities. Strong ties within and outside of work environment may well bring the high-investment effects of computer system. Overall, CEOs and managers should remember that important knowledge is still carried by the individuals of the rm and try to bring together these individuals to efciently share their historical knowledge and experience. Clearly, there are limitations to conclusions that can be drawn from this study. First, and perhaps most importantly, this study identied relationships, not causes. It is possible that something else was going on at the same time as knowledge sharing that led to some or all of the changes noted. For example, there may have been changes in the employee population from which we drew our samples, owing to layoffs and resignations. Respondents for the data collection may have used different psychological frames of reference in conducting the interviews. However, given that the project performance was measured by one person the top manager responsible for the project and given that interviews were conducted on a pre-planned survey protocol that underwent several stages of preliminary tests and pre-tests, the condence in attributing these results to the independent variables is bolstered. The method of conducting interviews personal interviews ensured that all the questions in the survey were completed by the respondents. Also, low response rate is a persistent problem in many survey type studies. Because this study was sanctioned by the head of the rm most managers of one department were included in the study, which made the response rate high, but due to the size of the department, the absolute sample size is still not large. Similarly, the self-reported perceptions of knowledge characteristics, tie strength, and ranking of knowledge elements, could have been cited as a limitation in survey type studies in lieu of objective departmental data. However, most objective data such as reports are based on these self-reported publications or reports and are not proofs of absolute objectivity. Moreover, these self-reported perceptions and evaluations on project success offer more information in terms of rst-hand experience by the persons in charge of each project. Most objective measures, such as time length, and budget offer very little information because these tend to be intrinsic to the nature of the project and not objective measures of performance. For example, a project involving military weaponry may be big in terms of budget and payment size but not necessarily successful if the project was conducted in an inefcient manner or resulted in wrong conclusions. This is why many aspects of the method in this study which may be considered a limitation in other survey studies may end up being strengths. Since this study is limited to a single rm, this nding can be considered unique to this rm, limited by the ability to measure tacitness. However, because this study was conducted in only one company, conceptual replications in other organizations will be required before concrete conclusions can be drawn. This study used only one rm by design to help eliminate across-rm differences and biases especially in evaluating project performance. However, the natural extension of this study is to resume the analysis with a broader sample

PAGE 140 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007

size, not only across the departments of a rm, but also to include projects from other rms in the same industry as well as other industries. It would be interesting to see if these results carry over to other similar rms in knowledge-intensive industries.

References
Allen, T.J. (1977), Managing the Flow of Technology: Technology Transfer and the Dissemination of Technological Information within the R&D Organization, MIT Press, Cambridge, MA. Argyres, N.S. and Silverman, B.S. (2004), R&D, organization structure, and the development of corporate technological knowledge, Strategic Management Journal, Vol. 25 Nos 8-9, pp. 929-58. Bain, J.S. (1959), Industrial Organization, Wiley, New York, NY. Barney, J.B. (1991), Firm resources and sustained competitive advantage, Journal of Management, Vol. 17 No. 1, pp. 99-120. Bechky, B.A. (1999), Creating shared meaning across occupational communities: an ethnographic study of a production oor, Proceedings of the 1999 Academy of Management Conference, Chicago, IL. Bierly, P. and Chakravarti, A. (1996), Generic knowledge strategies in the US pharmaceutical industry, Strategic Management Journal, Vol. 17, Winter, special issue, pp. 123-35. Birkinshaw, J., Nobel, R. and Ridderstrale, J. (2002), Knowledge as a contingency variable: do the characteristics of knowledge predict organizational structure?, Organization Science, Vol. 13 No. 3, pp. 274-89. Brown, J.S. and Duguid, P. (1991), Organizational learning and communities-of-practice: toward a unied view of working, learning and innovation, Organization Science, Vol. 2 No. 1, pp. 40-57. Brown, J.S. and Eisenhardt, K.M. (1998), Competing on the Edge Strategy as Structured Chaos, Harvard Business School Press, Boston, MA. Brown, S.L. and Eisenhardt, K.M. (1995), Product development: past research, present ndings, and future directions, Academy of Management Review, Vol. 20 No. 2, pp. 343-78. Brown, S.L. and Eisenhardt, K.M. (1997), The art of continuous change: linking complexity theory and time-paced evolution in relentlessly shifting organizations, Administrative Science Quarterly, Vol. 42 No. 1, pp. 1-34. Cohen, W.M. and Levinthal, D.A. (1990), Absorptive capacity: a new perspective on learning and innovation, Administrative Science Quarterly, Vol. 35 No. 1, pp. 128-52. Conner, K.R. and Prahalad, C.K. (1996), A resource-based theory of the rm: knowledge versus opportunism, Organization Science, Vol. 7 No. 5, pp. 477-501. Corcoran, K. and Fischer, J. (1987), Measures for Clinical Practice: A Sourcebook, The Free Press, New York, NY. Cronbach, L.J. (1984), Essentials of Psychological Testing, 4th ed., Harper & Row, New York, NY. Cummings, J.N. (2004), Work groups, structural diversity, and knowledge sharing in a global organization, Management Science, Vol. 50 No. 3, pp. 352-64. Cyert, R.M. and March, J.G. (1963), A Behavioral Theory of the Firm, Prentice-Hall, Englewood Cliffs, NJ. Dierickx, I. and Cool, K. (1989), Asset stock accumulation and sustainability of competitive advantage, Management Science, Vol. 35 No. 12, pp. 1504-11. Diesing, P. (1971), Patterns of Discovery in the Social Sciences, Aldine Atherton, Chicago, IL and New York, NY. Eisenhardt, K. and Brown, S. (1998), Competing on the Edge: Strategy as Structured Chaos, Harvard Business School Press, Boston, MA. Eisenhardt, K. and Galunic, D.C. (2000), Coevolving: at last, a way to make synergies work, Harvard Business Review, January-February, pp. 91-100. Eisenhardt, K. and Santos, F.M. (2001), Knowledge-based view: a new theory of strategy?, in Pettigrew, A., Thomas, H. and Whittington, R. (Eds), Handbook of Strategy and Management, Sage Publications, London, Thousand Oaks, CA and New Delhi, pp. 139-64.

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 141

Granovetter, M.S. (1973), The strength of weak ties, American Journal of Sociology, Vol. 78 No. 6, pp. 1360-80. Granovetter, M.S. (1982), The strength of weak ties: a network theory revisited, in Marsden, P. and Lin, N. (Eds), Social Structure and Network Analysis, Sage, Beverly Hills, CA, pp. 105-30. Grant, R. (1991), The resource-based theory of competitive advantage, California Management Review, Vol. 33 No. 3, pp. 114-34. Grant, R.M. (1996), Knowledge, strategy and the theory of the rm, Strategic Management Journal, Vol. 17 No. S2, pp. 109-22. Gupta, A.K. and Govindarajan, V. (2000), Knowledge ow within multinational corporations, Strategic Management Journal, Vol. 21 No. 4, pp. 473-96. Hansen, M. (1998), Combining network centrality and related knowledge: explaining effective knowledge sharing in multiunit rms, working paper, Harvard Business School, Boston, MA. Hansen, M. (1999), The search-transfer problem: the role of weak ties in sharing knowledge across organizational subunits, Administrative Science Quarterly, Vol. 44 No. 1, pp. 82-111. Hansen, M., Mors, M.L. and Lovas, B. (2005), Knowledge sharing in organizations: multiple networks, multiple phases, Academy of Management Journal, Vol. 48 No. 5, pp. 776-93. Hargadon, A. (1998), Knowledge brokers: a eld study of organizational learning and innovation, Proceedings of the 1998 Academy of Management Conference, San Diego, CA. Hargadon, A. and Sutton, R.I. (1997), Technology brokering and innovation in a product development rm, Administrative Science Quarterly, Vol. 42 No. 4, pp. 716-49. Helper, S., MacDufe, J.P. and Sabel, C. (1999), Pragmatic collaborations: advancing knowledge while controlling opportunism, working paper. Henderson, R. (1994), The evolution of integrative capability: innovation in cardiovascular drug discovery, Industrial and Corporate Change, Vol. 3 No. 3, pp. 607-30. Henderson, R. and Cockburn, I. (1994), Measuring competence? Exploring rm effects in pharmaceutical research, Strategic Management Journal, Vol. 15 No. 8, pp. 63-84. Henderson, R. and Cockburn, I. (1996), Scale, scope, and spillovers: the determinants of research productivity in drug discovery, RAND Journal of Economics, Vol. 27 No. 1, pp. 32-59. Imai, K., Nonaka, I. and Takeuchi, H. (1985), Managing the new product development process: how Japanese rms learn and unlearn, in Clark, K., Hayes, R. and Lorenz, C. (Eds), The Uneasy Alliance, Harvard Business School Press, Boston, MA, pp. 337-76. Jett, Q. (1999), All the right moves: linking organizational capabilities with competitive product moves in dynamic markets, working paper. Katz, R. and Tushman, M. (1981), An investigation into the managerial role and career paths of gatekeepers and project supervisors in a major R&D facility, R&D Management, Vol. 11 No. 3, pp. 103-10. Kim, K., Park, J.H. and Prescott, J. (2003), The global integration of business functions: a study of multinational businesses in integrated global industries, Journal of International Business Studies, Vol. 34 No. 4, pp. 327-44. Kogut, B. and Zander, U. (1993), Knowledge of the rm and the evolutionary theory of the multinational corporation, Journal of International Business Studies, Vol. 24 No. 4, pp. 625-45. Kogut, B. and Zander, U. (1995), Knowledge, market failure and the multinational enterprise: a reply, Journal of International Business Studies, Vol. 26 No. 2, pp. 417-26. Kogut, B. and Zander, U. (1996), What rms do? Coordination, identity, and learning, Organization Science, Vol. 7 No. 5, pp. 502-18. Krackhardt, D. (1992), The strength of strong ties, in Nohria, N. and Eccles, R.G. (Eds), Networks and Organizations: Structure, Form and Action, Harvard Business School Press, Boston, MA, pp. 216-39. Kwak, M. (2001), Searching for search costs, MIT Sloan Management Review, Vol. 42 No. 3, pp. 8-9.

PAGE 142 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007

Levin, D. and Cross, R. (2004), The strength of weak ties you can trust: the mediating role of trust in effective knowledge transfer, Management Science, Vol. 50 No. 11, pp. 1477-90. Liebeskind, J., Oliver, A., Zucker, L. and Brewer, M. (1996), Social networks, learning, and exibility: sourcing scientic knowledge in new biotechnology rms, Organizational Science, Vol. 7 No. 4, pp. 428-43. Lord, M. and Ranft, A. (1998), Transfer and sharing of local knowledge within the rm and entry into new international markets, Proceedings of the 1998 Academy of Management Conference, Chicago, IL. McEvily, B. and Zaheer, A. (1999), Bridging ties: a source of rm heterogeneity in competitive capabilities, Strategic Management Journal, Vol. 20, pp. 1133-56. Mowery, D., Oxley, J. and Silverman, B. (1996), Strategic alliance and interrm knowledge transfer, Strategic Management Journal, Vol. 17 special issue, Winter, pp. 77-91. Myers, S. and Marquis, D.G. (1969), Successful Industrial Innovations: A Study of Factors Underlying Innovation in Selected Firms, National Science Foundation, Washington, DC. Nelson, R.R. and Winter, S.G. (1982), An Evolutionary Theory of Economic Change, Harvard University Press, Cambridge, MA. Nunnally, J.C. (1978), Psychometric Theory, 2nd ed., McGraw-Hill, New York, NY. Okhuysen, G.A. and Eisenhardt, K. (2002), Integrating knowledge in groups: how simple formal interventions enable exibility, Organization Science, Vol. 13 No. 4, pp. 370-86. Park, Y. and Kim, S. (2005), Linkage between knowledge management and R&D management, Journal of Knowledge Management, Vol. 9 No. 4, pp. 34-44. Polanyi, M. (1962), Personal Knowledge: Towards a Post-critical Philosophy, Routledge & Kegan Paul, London (rst published in 1958). Porter, M. (1980), Competitive Strategy: Techniques for Analyzing Industries and Competitors, Free Press, New York, NY. Powell, W., Koput, K. and Smith-Doerr, L. (1996), Interorganizational collaboration and the locus of innovation: networks of learning in biotechnology, Administrative Science Quarterly, Vol. 41 No. 1, pp. 116-45. Prahalad, C.K. and Hamel, G. (1990), The core competence of the corporation, Harvard Business Review, Vol. 68, May-June, pp. 79-91. Prusak, L. (1999), Workshop on Creativity and Knowledge Creation, Reginald H. Jones Center for Management Policy, Strategy and Organization and the SEI Center for Advanced Studies in Management, Philadelphia, PA. Reagans, R. and Zuckerman, E.W. (2001), Networks, diversity, and productivity: the social capital of corporate R&D teams, Organization Science, Vol. 12 No. 4, pp. 502-17. Rogers, E. (1962), Diffusion of Innovations, Free Press, New York, NY. Rosenkopf, L. and Nerkar, A. (2001), Beyond local search: boundary-spanning, exploration, and impact in the optical disc industry, Strategic Management Journal, Vol. 22 No. 4, pp. 287-306. Rothwell, K. (1972), Administrative Issues in Developing Economies, Lexington Books, Lexington, MA. Sanchez, R. and Heene, A. (1997), Strategic Learning and Knowledge Management, Wiley, Chichester. Sanchez, R., Heene, A. and Thomas, H. (1996), Dynamics of Competence-based Competition: Theory and Practice in the New Strategic Management, Pergamon, Oxford and New York, NY. Simon, H. (1965), Administrative Behavior, Free Press, New York, NY. Stalk, G., Evans, P. and Shulman, L.E. (1992), Competing on capabilities: the new rules of corporate strategy, Harvard Business Review, Vol. 70, March-April, pp. 57-69. Sundberg, N.D. (1977), Assessment of Persons, Prentice-Hall, Englewood Cliffs, NJ. Szulanski, G. (1996), Exploring internal stickiness: impediments to the transfer of best practice within the rm, Strategic Management Journal, Vol. 17 No. S2, pp. 27-44.

VOL. 11 NO. 3 2007 JOURNAL OF KNOWLEDGE MANAGEMENT PAGE 143

Teece, D.J. (1996), Firm organization, industrial structure, and technological innovation, Journal of Economic Behavior and Organization, Vol. 31 No. 2, pp. 193-225. Teece, D.J. and Pisano, G. (1994), The dynamics capabilities of rms: an introduction, Industrial Corporate Change, Vol. 3 No. 3, pp. 537-56. Teece, D.J., Pisano, G. and Shuen, A. (1997), Dynamic capabilities and strategic management, Strategic Management Journal, Vol. 18 No. 7, pp. 509-33. Tripsas, M. (1997), Unravelling the process of creative destruction: complementary assets and incumbent survival in the typesetter industry, Strategic Management Journal, Vol. 18, special issue, Summer, pp. 119-42. Tsai, W. (2001), Knowledge transfer in intraorganizational networks: effects of network position and absorptive capacity on business unit innovation and performance, Academy of Management Journal, Vol. 44 No. 5, pp. 996-1004. Uzzi, B. (1996), The sources and consequences of embeddedness for the economic performance of organizations: the network effect, American Sociological Review, Vol. 61 No. 4, p. 674. Wernerfelt, B. (1984), A resource-based view of the rm, Strategic Management Journal, Vol. 5 No. 2, pp. 171-80. Winter, S.G. (1987), Knowledge and competence as strategic assets, in Teece, D.J. (Ed.), The Competitive Challenge: Strategy for Industrial Innovation and Renewal, Harper & Row, New York, NY, pp. 159-84. Zander, U. and Kogut, B. (1995), Knowledge and the speed of transfer and imitation of organizational capabilities: an empirical test, Organization Science, Vol. 6 No. 1, pp. 76-92.

About the author

Jina Kang is a faculty member of the Department of Industrial Engineering and the Graduate Program in Technology and Management (TAM) at Seoul National University (SNU). She received her PhD from the Anderson School of the University of California, Los Angeles (UCLA), MBA from the Wharton School of the University of Pennsylvania (UPenn), and BS in Management Science from Korea Advanced Institute of Science and Technology (KAIST). Her current research interests focus on the characteristics of knowledge and social networks in organizations and their impact on rm performance, and how innovators can appropriate from their innovations. Her broad research interests span knowledge, innovation, technology, and network theory. Jina Kang can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

PAGE 144 JOURNAL OF KNOWLEDGE MANAGEMENT VOL. 11 NO. 3 2007