reading analysis (graduate level)

Public Management Research Association Journal of Public Administration Research and Theory, Inc. Knowledge Sharing in Public Sector Organizations: The Effect of Organizational Characteristics on Interdepartmental Knowledge Sharing Author(syf $ Q Q L F N : L O O H P D Q G 0 D U F % X H O H Q s Source: Journal of Public Administration Research and Theory: J-PART , Vol. 17, No. 4 (Oct., 2007yf S S 6 Published by: Oxford University Press on behalf of the Public Management Research Association Stable URL: http://www.jstor.org/stable/25096342 Accessed: 03-10-2016 19:59 UTC JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms Public Management Research Association, Journal of Public Administration Research and Theory, Inc., Oxford University Press are collaborating with JSTOR to digitize, preserve and extend access to Journal of Public Administration Research and Theory: J-PART This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms JPART 17:581-606 Knowledge Sharing in Public Sector Organizations: The Effect of Organizational Characteristics on Interdepartmental Knowledge Sharing Annick Willem Ghent University Marc Buelens Ghent University and Vlerick Leuven Gent Management School ABSTRACT Public sector organizations are mainly knowledge-intensive organizations, and to exploit their knowledge, effective knowledge sharing among the different departments is required. We focus on specific characteristics of public sector organizations that increase or limit interdepartmental knowledge sharing. Three types of organization-specific coordination mechanisms directly influence knowledge sharing between departments. Organizations are also characterized by members' social identification and trust, which in the absence of power games are assumed to create a knowledge-sharing context. Data are collected by a questionnaire survey in the public sector. The sample consists of 358 cooperative epi sodes between departments in more than 90 different public sector organizations. Struc tural equation modeling reveals the importance of lateral coordination and trust. The combination of power games and informal coordination seems to be remarkably beneficial for knowledge sharing. Furthermore, compared with other public sector organizations, gov ernment institutions have organizational characteristics that are less beneficial for knowledge sharing. INTRODUCTION Knowledge is generally seen as one of the most, if not the most, important asset in organizations that should be carefully managed (Argote, McEvily, and Reagans 2003; Teece 1998yf + R Z H Y H U N Q R Z O H G J H P D Q D J H P H Q W D V Q H Z P D Q D J H U L D O S U D F W L F H D Q G D V D n academic research topic has not yet entered the public sector literature, with exceptions of studies in certain public services such as health care (Bate and Robert 2002yf 7 K L V L s surprising, knowing that public sector organizations often have as their main activity developing and providing knowledge (Luen and Al-Hawamdeh 2001yf D Q G K H Q F H F D Q E e Address correspondence to the author at [email protected]. doi: 10.1093/jopart/mul021 Advance Access publication on January 4, 2007 ? The Author 2007. Published by Oxford University Press on behalf of the Journal of Public Administration Research and Theory, Inc. All rights reserved. For permissions, please e-mail: [email protected] This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 582 Journal of Public Administration Research and Theory classified as knowledge-intensive organizations. All organizations are to some extent knowledge intensive (Alvesson 1993yf + R Z H Y H U V R P H R U J D Q L ] D W L R Q V K D Y H N Q R Z O H G J H D s their core product, provide knowledge to the public as their main activity, or have mainly knowledge workers, that is, experts developing and providing knowledge (Starbuck 1992yf . This is characteristic of most public sector organizations. Within the growing body of knowledge management literature, increasing attention is being paid to the role of organization structure in knowledge sharing (Grover and Davenport 2001yf , Q W K L V D U W L F O H Z H I R F X V R Q N Q R Z O H G J H V K D U L Q J G X U L Q J F R R S H U D W L Y H H S i sodes between departments. A cooperative episode here refers to any kind of cooperation between one or more people from two departments. Knowledge sharing is defined as "the process through which one unit is affected by the experience of another" (Argote et al. 2000, 3yf 7 K L V Y L H Z R Q N Q R Z O H G J H V K D U L Q J L Q Y R O Y H V P R U H W K D Q V L P S O \ D F T X L U L Q J R U W U D Q s mitting knowledge from one party to another but is a process of exchanging and processing knowledge in a way that knowledge of one unit can be integrated and used in another unit. Knowledge is a very broad concept, and there is no consensus on its boundaries and definition. In this article, we limit knowledge to practical and experience-based knowledge or know-how, that is, "practical understanding that enables a firm to perform various operations" (Sanchez and Heene 1997, 178yf 6 X F K N Q R Z O H G J H L V G X H W R L W V H P E H G G H d and tacit character, distributed and hidden away in individuals' cognition and behavior and in organizational units (Birkinshaw, Nobel, and Ridderstrale 2002; Szulanski 2000yf ' e partmentalization is a natural and unavoidable barrier to the sharing of such knowledge, preventing one department from benefiting from the experiences of another (Argote et al. 2000yf . Organizations can develop proper structures to leverage this knowledge sharing be tween departments (Teece 1998yf 7 K H S U R E O H P R I G H V L J Q L Q J D Q R U J D Q L ] D W L R Q W K D W R S W L P L ] H s knowledge sharing remains unsolved, but several studies have shed light on the issue and reveal insight into the relevant influencing factors (Easterby-Smith and Lyles 2003yf 2 Q e important facilitator of knowledge sharing between departments is the coordination that exists between departments (Grant 1996yf & R R U G L Q D W L R Q P H F K D Q L V P V I D F L O L W D W H L Q W H U D F W L R Q , cooperation, and the exchange of resources, such as knowledge, between departments (Galbraith 1973; Grandori 1997yf + H Q F H Z H D U H L Q W H U H V W H G L Q K R Z W K H P H F K D Q L V P V R f coordination impact knowledge sharing during cooperative episodes. Note that coordina tion mechanisms are different from cooperative episodes because the former are facilita tors of the latter; in other words, a cooperative episode can be made possible by one or more coordination mechanisms. Furthermore, knowledge management literature has in dicated that several other organizational characteristics also influence the possibilities for knowledge sharing between units, such as trust, power games, and social identification (Andrews and Delahaye 2000; Cabrera and Cabrera 2002; K?ser and Miles 2002yf 7 K H V e characteristics shape the context for interdepartmental knowledge sharing. Therefore, we also take into account the effect that these characteristics have on the relationship between coordination mechanisms and knowledge sharing. We first look at the particularities of organization design in public sector or ganizations. The article then continues with a description of the coordination mechanisms and in the next section develops some propositions. The next section describes our method of data collection. The following sections provide the analysis and discussion of our findings. We conclude the article with the contributions and limitations of our findings. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 583 ORGANIZATIONAL CHARACTERISTICS OF PUBLIC SECTOR ORGANIZATIONS Our study focuses on the relationship between knowledge sharing and organizational characteristics in public sector organizations. Although the design of public sector organ izations has been studied for several decades, it remains a controversial issue. The specific values in public sector organizations have strong implications for organization design (Hood 1991; Posner and Schmidt 1996yf 7 K H U H O D W L Y H O \ K L J K L P S R U W D Q F H R I Y D O X H V V X F h as honesty, fairness, and equity compared to more economic and parsimonious values, such as cost control and goal orientedness, demands a fundamentally different design. Hence, although we can identify a limited number of generic types of organization design (Mintzberg 1989yf D Q G F R R U G L Q D W L R Q P H F K D Q L V P V * U D Q G R U L \f, particular organization designs and coordination mechanisms are more appropriate and more frequently present in public sector organizations. Mintzberg (1989yf I R U L Q V W D Q F H V D \ V W K D W W K H P D F K L Q e bureaucracy as an organization type is more common in governmental organizations and that such a type is characterized by the standardization of processes. A focus in the organization design choices on processes rather than on output is required to support public sector values (Harmon and Mayer 1986; Hood 1991yf $ W I L U V W J O D Q F H O L W H U D W X U H R Q S X E O L c private sector differences seems to support the existence of more bureaucratic character istics in public sector organizations (Boyne 2002; Perry and Rainey 1988; Rainey 1997yf . However, empirical evidence about the existence of bureaucracy and formalization in public sector organizations is not fully straightforward. The review by Boyne (2002yf V K R Z s mixed results. Only six out of 11 studies confirm the bureaucracy hypothesis, and there is empirical support for arguments that rules serving as control and coordination mechanisms are just as important in private as in public organizations (Boyne 2002yf 7 K H D Q D O \ V L V R f Boyne (2002yf D O V R V K R Z H G W K D W L Q W K U H H R I I L Y H V W X G L H V R U J D Q L ] D W L R Q D O F R P P L W P H Q W Z D s lower in public organizations. This was confirmed in a study by Goulet and Frank (2002yf . Rainey and Bozeman (2000yf S R L Q W H G R X W W K H P H W K R G R O R J L F D O O L P L W D W L R Q V D Q G F R Q F H S W X D l difficulties in convergent findings of different studies. However, they also emphasized that aggregating findings of studies to compare public and private organizations provides useful insights. Hence, although the evidence is weak, the differences in, for instance, commit ment and process orientation, should not be neglected. The emphasis on processes has led in some public sector organizations to dysfunc tional bureaucracies and problems with so-called red tape, referring to lengthy procedures and huge amounts of documents accompanying the many procedures. New public man agement emphasizes more economic and parsimonious values, that is, values oriented toward frugality of resource use such as cost-effectiveness and output control (Hood 1991yf 7 K L V Q H Z S X E O L F P D Q D J H P H Q W K D G W K H L Q W H Q W L R Q W R L Q F U H D V H W H D P Z R U N O L P L W W K e number of rules, and obtain more value-driven management and decentralization in public sector organizations (Osborne and Gaebler 1993yf K R Z H Y H U H P S L U L F D O H Y L G H Q F H S U R Y H G W K D t these aims were not always reached (Hood and Peters 2004yf , Q D G G L W L R Q X V L Q J P R U e private sector instruments might be in conflict with the values of the public sector (Moe and Gilmour 1995yf 2 O V H Q \f, for instance, explains the benefit of bureaucratic organizational forms for public organizations. Therefore, there is a need to study and develop distinctive organization design principles for public sector organizations (Fottler 1981yf . Furthermore, there is not just one organization design approach to public sector organizations because there is no single type of public sector organization. Public sector This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 584 Journal of Public Administration Research and Theory organizations differ from private sector organizations in the ambiguity of goals, the difference in environment, and the difference in political influences?differences that in fluence the design of public sector organizations; however, even within the public sector, these aspects vary greatly (Heffron 1989yf 3 H U U \ D Q G 5 D L Q H \ \f emphasized that the public sector is a multidimensional concept with many dimensions that determine the extent to which an organization is considered public. Relevant dimensions are ownership, funding of the organization, and control (Bozeman 1987yf 2 W K H U G L P H Q V L R Q V L Q F O X G H L n terest (benefits for individuals or for the communityyf D F F H V V W R I D F L O L W L H V D Q G D J H Q F y (whether or not the organization is acting as an agent for the communityyf % H Q Q D Q G * D X s 1983yf 7 K L V P X O W L G L P H Q V L R Q D O F K D U D F W H U R I W K H S X E O L F V H F W R U F R Q F H S W E U L Q J V D E R X W D F R Q W L n uum of public sector organizations that are, to a greater or lesser degree, public (Boyne 2002; Fottler 1981yf ) R U W K H S X U S R V H R I R X U V W X G \ Z H G L V W L Q J X L V K E H W Z H H Q W K U H H W \ S H V R f public sector organizations along this continuum, namely, government institutions (i.e., public administration and governmentsyf S X E O L F V H F W R U L Q V W L W X W L R Q V D Q G V W D W H H Q W H U S U L V H V . Hence, our categorization of organizations is based on the dimensions of access, interest, and agency as defined by Benn and Gaus (1983yf * R Y H U Q P H Q W L Q V W L W X W L R Q V D F W D V D J H Q W V I R r the interest and access for the community. These institutions are the federal, regional, and local governments (municipalitiesyf 3 X E O L F V H F W R U L Q V W L W X W L R Q V G R Q R W D F W D V D J H Q W V I R U W K e community but provide access to facilities and work for the interest of the whole commu nity. Examples of public sector institutions are schools, public hospitals, public prisons, and several other nonprofit organizations providing services to the public. In the state enterprises, access and interest are only to a limited extent communal. These state enter prises, such as postal services, operate more similar to private enterprises. We classify government institutions as most public, public sector organizations as next most public, and state enterprises as least public. Although the importance of organization design and coordination mechanisms for knowledge sharing is recognized (Grant 1996; Van den Bosch, Volberda, and de Boer 1999yf W K H Q X P E H U R I V W X G L H V R Q W K H U H O D W L R Q V K L S E H W Z H H Q F R R U G L Q D W L R Q D Q G N Q R Z O H G J e sharing is very limited, and none of the studies have this particular relationship as its core research aim. Moreover, there are no authors who assess this relationship in public sector organizations. This article contributes to our understanding of knowledge sharing in public sector organizations, taking into account their specific organizational characteristics. In particular, we further develop insights into how organizational characteristics affect in terdepartmental knowledge sharing. Hence, we seek answers to the following questions: "What kind of bridges between departments in public sector organizations are required to make interdepartmental sharing of knowledge possible?" and "Which organizational char acteristics are dominant in knowledge sharing?" We examine the literature on public sector organizations, knowledge sharing, and organization theory to increase our understanding of what drives knowledge sharing in three kinds of public sector organizations. Theoretical insights are further studied based on data collected using a questionnaire distributed in a broad range of public sector organizations. COORDINATION MECHANISMS AND INTERDEPARTMENTAL KNOWLEDGE SHARING In reviewing organization theory literature, we identified three types of interunit coordi nation mechanisms based on the level of formalization (Ghoshal, Korine, and Szulanski This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 585 1994; Martinez and Jarillo 1989yf D Q G W K H S U R J U D P P H G F K D U D F W H U $ U J R W H 9 D Q ' e Ven, Delbecq, and Koenig 1976yf R I F R R U G L Q D W L R Q : H W K X V G L V W L Q J X L V K H G E H W Z H H Q I R U P D l and informal coordination. Formal coordination can be further divided into two categories based on the extent to which coordination plans and programs the tasks in advance of the task execution. Hence, first, there are formal systems that consist of any kind of coordi nation that is planned and formally established, such as formal procedures, rules, manuals, and formal processes. Second, there is lateral coordination that is also formal but not planned in advance. This coordination is used when needed during task execution. Exam ples are teamwork, liaison roles, task groups, and mutual adjustment. The third category includes all informal coordination. Mintzberg (1989yf U H I H U U H G D O V R W R W K H V W D Q G D U G L ] D W L R Q R I Y D O X H V D V D N L Q G R I S U o grammed coordination mechanism. Values that are part of an organization's or group's culture can program organizational behavior; through this programming, coordination is achieved. However, values do not directly make cooperative episodes and interdepartmen tal knowledge sharing possible. It is the extent to which people identify with the values that is important and that can help in creating a favorable cooperative and knowledge-sharing context. Kogut and Zander (1996yf D U J X H G W K D W V R F L D O L G H Q W L W \ S O D \ V D S L Y R W D O U R O H L Q R Y H r coming problems of communication among the specialists created through the division of labor. In explicating this role, they identified the influence that identity exerts on attribution (translating individual self-interest into collective interestyf D Q G F R R U G L Q D W L R Q W K U R X J K F R n vergent expectations and developing shared interpretationsyf $ V K I R U W K D Q G 0 D H O ; Maguire, Phillips, and Hardy 2001yf , G H Q W L I L F D W L R Q I X U W K H U U H V X O W V L Q S U R V R F L D O E H K D Y L R r (Jarvenpaa and Staples 2001yf D Q G G H Y H O R S V W U X V W 0 F ( Y L O \ 3 H U U R Q H D Q G = D K H H U \f. Identifying with cultural values also reduces opportunistic behavior such as the kind of behavior existing in power games. We include this prosocial behavior through social identity, trust, and (lack ofyf S R Z H U J D P H V D V F R Q W H [ W X D O R U J D Q L ] D W L R Q D O F K D U D F W H U L V W L F V . Hence, there are two kinds of organizational characteristics influencing interdepartmental knowledge sharing. One group, the coordination mechanisms, literally coordinates peo ple's tasks and makes cooperative episodes possible. This group includes the formal systems, lateral coordination, and informal networks. The other group creates context by tuning expectations and by establishing a willingness to engage in cooperative episodes. This group refers to identification, trust, and a lack of power games. Although each of the three types of coordination mechanisms and the contextual characteristics can be present in public sector organizations, the literature suggests that there might be a higher use of formal systems to cope with the equity values (Heffron 1989; Hood 1991; Moe and Gilmour 1995yf D Q G O R Z H U O H Y H O V R I L G H Q W L I L F D W L R Q L Q S X E O L F V H F W R r organizations, compared to private sector organizations, due to lower commitment (Boyne 2002; Moon 2000yf , G H Q W L I L F D W L R Q Z L W K W K H R U J D Q L ] D W L R Q D Q G L W V J R D O V L V D Q L P S R U W D Q W G i mension of commitment (Moon 2000yf 2 Q H S R V V L E O H H [ S O D Q D W L R Q R I O R Z H U L G H Q W L I L F D W L R n in public sector organizations might be the impossibility of employees in public sector organizations of identifying with the final service delivered and the impossibility of ob serving a relationship between employees' contributions and the performance of the or ganization (Boyne 2002; Perry and Porter 1982yf 7 K L V H I I H F W Z L O O E H P R V W O \ S U H V H Q W L Q W K e "pure" public sector organizations or the government institutions. Organization theory and knowledge management literature teach us that lateral co ordination and informal networks are considered to be better fit for knowledge sharing for several reasons. First, formal systems fully plan the coordination between units, and hence, This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 586 Journal of Public Administration Research and Theory the possibilities for knowledge sharing are limited to what was planned (Grant 1996yf 6 \ s tems can, for instance, determine the knowledge flow in the different steps in a sequentially organized delivery process. Although this type of coordination is considered to have a low cost, it has limited potential for enhancing knowledge sharing (Grant 1996yf 6 \ V W H P s require some kind of stability or analyzability and, consequently, possibilities for planning. Therefore, systems are unfit when flexibility in the knowledge-sharing process is required. The lateral and more networklike coordination types, which are not strictly planning activities in advance, do not have this drawback and might therefore result in more flexible and timely knowledge sharing. The learning organization literature has emphasized the need for more lateral coordination mechanisms to allow learning and knowledge sharing because of the flexibility and decentralized character of this coordination (Ayas and Zeniuk 2001; Duncan and Weiss 1979; Senge 1990yf + D \ Q H V \f also argued in favor of team based professional work instead of using standardized systems for knowledge management in the public sector. The informal character of informal coordination results in even more flexibility, especially in crossing the formal boundaries in the organization (Gargiulo and Benassi 2000; Hansen 1999yf . Second, informal and lateral coordination result in more intense communication in the cooperative episodes than formal systems and, hence, result in more opportunities for knowledge sharing (Nidumolu 1996yf W K X V S R W H Q W L D O O \ U H V X O W L Q J L Q D E U R D G H U R U J D Q L ] D W L R n wide diffusion of knowledge (Tsai 2002yf , Q I R U P D O F R R U G L Q D W L R Q R I I H U V H Y H Q P R U H R p portunities to cross-departmental boundaries because it is less bound to the formally established links between departments than the lateral coordination. Third, it is generally accepted in the knowledge management literature that "sticky," embedded, and implicit knowledge is difficult to share (Hansen 1999; Szulanski 2000yf . This embedded and implicit character of individual knowledge makes hierarchy, also a formal system, unfit as a coordination mechanism (Grant 1996yf + D Q V H Q \f explains that informal coordination is best suited for the sharing of implicit knowledge, and Galbraith (1973yf P H Q W L R Q H G W K D W O D W H U D O F R R U G L Q D W L R Q F D Q G H D O E H W W H U Z L W K J U H D W H U F R P S O H x ity (as is the case with the sharing of sticky knowledgeyf W K D Q I R U P D O V \ V W H P V . Fourth, the more informal the coordination, the more an open atmosphere is present and the more people will be willing to put out the effort to share their knowledge. Bate and Robert (2002yf H P S K D V L ] H G W K H Q H H G I R U Y R O X Q W D U \ Q D W X U D O D Q G V S R Q W D Q H R X V S H U V R Q D O Q H t works with high levels of personal connectivity and social identity and low levels of management control to allow knowledge sharing in public sector organizations. Other authors have also emphasized the need for an open atmosphere, with a lack of control and high levels of sociability, to allow spontaneous and voluntary knowledge sharing (Andrews and Delahaye 2000; Constant, Kiesler, and Sproull 1994; Jarvenpaa and Staples 2001yf , Q V X F K H Q Y L U R Q P H Q W V W U X V W L V R I W H Q S U H V H Q W D Q R W K H U F R Q G L W L R Q W K D W H Q F R X U D J H s knowledge sharing. However, a unit strongly engaged in an informal network can perceive inflexibility to behave independently because of the many social obligations (Burt 1992; Gargiulo and Benassi 2000; Hansen 1999yf 1 H W Z R U N P H P E H U V I H H O R E O L J H G W R F R R S H U D W H Z L W K R W K H r network members instead of with nonmembers and to collect knowledge predominantly within the network. Hence, informal coordination based on social instead of formal rela tionships can have an ambiguous effect on knowledge sharing. Although, on the one hand, informal coordination provides the opportunity and context for knowledge sharing cross ing formal boundaries, on the other hand, it creates new informal boundaries. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 587 To summarize, the nonplanned and nonformal character of coordination results in more intense knowledge sharing in the cooperative episodes between departments. Hence, formal systems are less effective for leveraging knowledge sharing, especially when these substitute lateral coordination. We therefore expect less knowledge sharing when formal systems are used. As mentioned, there is some support in the literature for the more intensive use of formal systems in public organizations. Consequently, we expect, in general, low levels of knowledge sharing in public sector organizations. Considering the formal versus informal character of coordination, we can conclude that although neither of the two is preferable because both have their limitations, the limitations are clearly dif ferent. Therefore, we suggest that informal and lateral coordination should complement each other for intense knowledge sharing in cooperative episodes. This brings us to the following four propositions. Pia The more use of formal systems, the less knowledge sharing in cooperative episodes. Plb The more use of lateral coordination, the more knowledge sharing in cooperative episodes. Plc The more use of informal coordination, the more knowledge sharing in cooperative episodes. Pld There will be an interaction between lateral coordination and informal coordination, such that in the presence of both the positive effect on knowledge sharing in cooperative episodes will be stronger. The positive effects of the contextual organizational characteristics on knowledge sharing, such as trust and identification, are more explicitly recognized in the literature. Alvesson (2000, 2001yf . R J X W D Q G = D Q G H U \f, Robertson, Scarbrough, and Swan (2003yf D Q d Robertson and Swan (2003yf K D Y H H P S K D V L ] H G W K H L P S R U W D Q F H R I V R F L D O L G H Q W L I L F D W L R Q L n a group or in the organization to leverage knowledge sharing. The assumed lower level of identification in government institutions will cause lower levels of knowledge sharing in these organizations compared to other public sector organizations. The importance of trust as a driver of knowledge sharing has been most widely recognized (Adler 2001; Andrews and Delahaye 2000; Ciborra and Andreu 2001; De Cremer, Snyder, and Dewitte 2001; McEvily, Perrone, and Zaheer 2003; Newell and Swan 2000yf 7 K U H H S U R S R V L W L R Q V R Q W K e role of trust and social identity are formulated. P2a The more trust in the other parties, the more knowledge sharing in the cooperative episode. P2b The more identification with the organization, the more knowledge sharing in the cooperative episodes. P2c The more public the organization, the less identification and, thus, the less knowledge sharing. The negative effect of power games is also recognized (Husted and Michailova 2002yf . Power games refer here to the unjustified use of power for personal aims, for instance, to This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 588 Journal of Public Administration Research and Theory enrich oneself or one's department or to increase one's control in the organization. Such power games can be used to make one's knowledge acceptable and used by others (Berman Brown and Woodland 1999; March and Simon 1958yf R U I R U K R D U G L Q J R Q H V N Q R Z O H G J e (Husted and Michailova 2002; Scarbrough 2003yf 7 K H Q H W Z R U N O L W H U D W X U H S R L Q W V D O V R D t some more drawbacks of informal coordination. Its personal nature inhibits the risk of abuse of power. This can be based on knowing the relationships between the people in the network (Krackhardt 1990yf R U R Q S H U V R Q D O D Q G L Q I R U P D O G L U H F W F R Q W D F W V W K D W J L Y H W K e possibility for opportunistic behavior (Adler and Kwon 2002yf 7 K H U H I R U H L W L V X V H I X O W o assess whether or not a cooperative episode that is coordinated by informal coordination is also facing power games. In such a situation, the effect of informal coordination on knowledge sharing might no longer be positive. Thus, informal coordination can have a knowledge brokerage role or a power-abuse role. O'Toole and Meier (2004yf D U J X H G W K D t public sector networks are especially vulnerable to power games and abuse of networks for nonoptimal resource distribution because of the involvement of politics in determining the organization's (publicyf J R D O V 7 K U H H P R U H S U R S R V L W L R Q V D U H D G Y D Q F H G E D V H G R Q W K H D V V X P H d role of power games in knowledge sharing in the public sector. P3a The more power games, the less knowledge sharing in the cooperative episodes. P3b The more public the organization, the more power games and, thus, the less knowledge sharing in the cooperative episodes. P3c The fewer power games in the informally coordinated cooperative episodes, the more knowledge sharing in these cooperative episodes. METHOD Data were collected through a questionnaire survey among Belgian public sector workers. Our unit of analysis was a cooperative episode between two departments in a public sector organization. The population thus consisted of all interdepartmental cooperative episodes within Belgian public sector organizations. The unit of measurement is individuals an swering the questionnaire. Hence, we selected individuals who could answer the question naire for a specific cooperative episode. We started with a sample of more than 90 different public sector organizations with 358 respondents, answering the questionnaire for as many different interdepartmental cooperative episodes, selected based on a nonrandom snowball selection method. Students of Master of Public Administration courses selected respond ents in their own organization or an organization with which they had frequent contact. Although random samples are more preferable, there was no sample selection bias toward certain types of cooperative episodes. On the contrary, the selection method resulted in a diverse sample of organizations, respondents, and cooperative episodes. The respondents worked in public sector organizations in a broad sense, including respondents from government institutions (39.1yb \f, such as municipal administrations and the Flemish educational department; from public sector institutions (41.1yb \f, including educational institutions, such as public schools (10.9yb \f, and medical and care organizations, such as public hospitals (6.4yb \f; and from state enterprises (19.8yb \f, such as public bus transporta tion service and water supply service. Our sample thus included a large group of all three types of public sector organizations, although more government institutions and less public This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 589 sector institutes were included relative to their share in the population of public sector organizations in Belgium, which was, respectively, 27.4 and 56.2yb 1 2 6 6 \f. The number of respondents per organization varied, with most organizations having 1?10 respondents (27yb K D Y L Q J R Q O \ R Q H U H V S R Q G H Q W S H U R U J D Q L ] D W L R Q \f, five large organizations having between 11 and 24 respondents, and one large government institute (counting over 13,000 employeesyf K D Y L Q J U H V S R Q G H Q W V I U R P G L I I H U H Q W G H S D U W P H Q W V $ O O V L [ R U J D Q L ] D W L R Q s that have more than 10 respondents were government institutes because these were rather concentrated with a small number of very large organizations. The respondents were active in a broad range of functions and on different levels in the organization. The labeling of hierarchical classifications of functions differs in the different public sector organizations; therefore, we asked respondents to compare theirs with the more common and uniformly used classification in Belgium of nonmanaging, junior management, middle management, and senior management. A large number of our respondents worked in nonmanaging functions (44yb \f, reflecting the real situation in the Belgian public sector where many public sector functions are nonmanaging functions. For example, for the Belgian Federal Government, 63yb R I W K H I X Q F W L R Q V D U H Q R Q P D Q D J L Q J I X Q F W L R Q V * R Y H U Q P H Q W \f. There were 20.9yb P L G G O H P D Q D J H U V \b junior managers, 5.3yb V H Q L R U P D Q D J H U V D Q G \b failed to complete this question. There was also a balanced spread over male and female respondents with, respectively, 52yb P D O H D Q G \b female respondents, almost reflecting the real spread of 51yb I H P D O H V D Q G \b male employees in the public sector at that time in Belgium (NOSS 2005yf . We asked the respondents to identify a cooperative episode between their department and one other department of the same organization and to briefly describe this cooperative episode. These cooperative episodes varied and included internal client-supplier relation ships, cooperative episodes to accomplish common tasks and projects, interactions to provide advice and explanations, and interactions to plan and coordinate tasks. Some specific examples given were as follows: "there is daily intensive cooperation necessary between several individuals of the two departments because we take care of the people in the prison and the other department administrates the files of those people," "my depart ment, the hospital's pharmacy, cooperates with the hospital's accounting department to monitor our performance," "developing together an educational plan for each student that follows a part-time education at our department and a part-time education at the other department," "exchanging information on psychiatric patients that come to our department for occupational therapy," and "working out answers on policy questions from the min isters in workgroups or workshops." All items in the questionnaire (see Appendix 1 for the questionnaire itemsyf K D G W R E H D Q V Z H U H G I R U W K H F R R S H U D W L Y H H S L V R G H D Q G W K L V Z D V L Q G L F D W H d in capital letters on the first page of the questionnaire. On the same page of the question naire, it was also indicated in capital letters that the respondents should only consider practical work-related knowledge in the form of work experiences, work methods, and know-how about their tasks. In the questionnaire wording, we used mainly the word knowledge but also information, experiences, and ideas, which were considered synonyms by the respondents. This was confirmed by the high Cronbach's alpha score of the knowl edge-sharing scale. Hence, the knowledge sharing measured was limited to the sharing of know-how with the selected other department during a cooperative episode. There was no focus on particular tasks, thus including any know-how that was related to a task for which the cooperative episode existed or related to any other tasks that the departments performed in the past or future. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 590 Journal of Public Administration Research and Theory As mentioned in the introduction of this article, knowledge sharing is a process that goes beyond transmitting knowledge; it also includes processing knowledge and using the knowledge in the own unit. The exchange of experiences and know-how between depart ments is thus only one part of knowledge sharing because it does not result in the use ful application of shared knowledge. Therefore, we chose to measure the intensity of knowledge sharing and the actual use of the shared knowledge, that is, effectiveness of knowledge sharing. Knowledge-sharing intensity was measured by six new items inquiring about the extent to which knowledge sharing was initiated by the exchange of knowledge between the departments. Effectiveness of knowledge sharing was based on a combination of three items of the scale from Wathne, Roos, and von Krogh (1996yf D Q G W Z R R I W K H I R X r items of the scale from Becerra-Fernandez and Sabherwal (2001yf 7 K H V H L W H P V P H D V X U H d the extent to which the knowledge sharing resulted in new insights, ways of working, and greater effectiveness of working. Hence, these two dependent variables formed a proxy for our theoretical knowledge-sharing concept, that is, the process in which one unit is affected by the experiences of another unit. We measured the following as independent variables: three coordination variables (formal systems, lateral coordination, and informal coordinationyf W K U H H F R Q W H [ W X D O R U J a nizational variables (power games, trust, and identificationyf D Q G R Q H F R Q W U R O Y D U L D E O e (incentivesyf 6 H Y H U D O V F D O H V Z H U H E D V H G R Q H [ L V W L Q J T X H V W L R Q Q D L U H V 3 R Z H U J D P H V Z H U e measured using five items based on Devos, Vanderheyden, and Van den Broeck (2001yf , measuring "to what extent relationships between units are strained because of power games and favoritism." Identity was measured using a scale developed by Wagner and Moch (1986yf E X W R Q H R I W K H I L Y H L W H P V Z D V H [ F O X G H G W R L Q F U H D V H U H O L D E L O L W \ R I W K H V F D O H 7 K e scale measured the identification with the organization, in particular "the willingness of people to sacrifice for the unit and the organization's goals." Trust (i.e., believing that the other party in the cooperative episode is reliable and trustworthyyf Z D V P H D V X U H G X V L Q J W K e scale of McAllister (1995yf R Q H L W H P Z D V H [ F O X G H G I R U W K H V D P H U H D V R Q D V L Q W K H L G H Q W L W y scale. The items were also slightly adapted to refer to trust between units instead of between individuals. Lateral coordination (i.e., the use of task forces, interunit commit tees, and interdepartmental meetings, discussions, and coordinatorsyf Z D V P H D V X U H G X V L Q g the four items on horizontal interunit coordination of Miller and Droge (1986yf Z L W K W Z o additional items. We developed a new scale for formal systems including four items on the use of systems (i.e., use of procedures, process descriptions, reports, and rulesyf W K U H H R Q W K e use of hierarchy as a coordination mechanism (i.e., impact of supervisors and unit heads on interdepartmental cooperative episodes and task executionyf , Q I R U P D O F R R U G L Q D W L R Q Z D s measured using five newly developed items. This scale indicated "whether people had their own network of friends in the organization and were using this network professionally." Finally, we also measured incentives because organizations evaluating their employees on teamwork and strong cooperative behavior might obtain more knowledge sharing through this, whatever the kind of coordination used. Therefore, we included a variable for incen tives measured by six items on "the extent to which cooperative behavior and also knowl edge sharing were stimulated in the organization through including teamwork and knowledge sharing in the employees evaluation processes and by making teamwork and knowledge sharing an explicit organizational goal." All items were measured on a five-point Likert-type scale ranging from 1 (=fully disagreeyf W R I X O O \ D J U H H \f. Cronbach's alpha was used as a measure of reliability, providing a lower bound for the reliability of itemized scales (Nunnally 1978yf $ O O E X t This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 591 one variable had an alpha greater than .70, thus providing an adequate level of reliability (see Appendix 2yf 7 K H R Q H Y D U L D E O H W K D W G L G Q R W U H D F K W K H F U L W H U L R Q V F R U H G R Q O \ P D U J L Q D O O y below the standard with an alpha of .69. Through factor analysis, unidimensionality was assessed and found to be adequate. The self-selection of the cooperative episodes can maintain a risk of bias. Although it was not asked to select frequent or successful co operative episodes, respondents might have had a tendency to select successful cooperative episodes. Conversely, respondents might remember especially the problematic cooperative episodes and select such episodes for completing the questionnaire. Most likely, however, respondents might have selected the cooperative episode that was most recent or even currently happening. In the latter case, we do not expect any bias in terms of success of the cooperative episode. Our variables have a normal spread in their values, also indicating absence of bias. A bias toward positive cooperative episodes could have resulted in com mon method error variance. Self-report measures, such as questionnaires, have a risk of common method error variance caused by measuring the dependent and independent variable at the same time by the same person and with the same instrument, without any possibility for cross-verification. The risk of common method variance was tested using the one-factor test of Hartman (Podsakoff and Organ 1986yf 7 K L V W H V W L Q G L F D W H V Z K H W K H U R r not the variables in one-factor analysis all load on one factor because such a situation would be an indication of common method variance. Applying the test to our data indicated three instead of one factor and, thus, did not indicate a common method variance problem. Other risks inherent to the method chosen here are single-source and self-report bias. A multisource method or other forms of triangulation are often suggested as a solution. However, Spector (1994, 2006yf F O D L P V W K D W H Y H Q W K H D O W H U Q D W L Y H V G R Q R W Q H F H V V D U \ U H V X O t in higher validity, but we should be cautious with self-report and single-source measure ment. Nevertheless, these techniques are useful in new organizational behavior fields as is the case here. The self-selection of cooperative episodes resulted also in a varied sample. We consider this as a strength of our study because relationships found are not context specific but clearly general for a diverse range of cooperative episodes in different public sector organizations in Belgium. RESULTS The correlations between the variables and mean scores are listed in Appendix 2. These descriptive values indicated the low use of incentives in the public sector organizations and the existence and use of informal coordination. Power games were present but not dom inant, and both knowledge-sharing intensity and knowledge-sharing effectiveness were moderately present. The intercorrelations indicated the negative relationship between power games and most of the other variables, except with informal coordination and formal systems. Lateral coordination positively correlated with all variables, except with power games where the relationship was negative. The intercorrelation between lateral coordi nation and incentives was high (.54yf D V H [ S H F W H G E H F D X V H L Q F H Q W L Y H Z D V S D U W O \ P H D V X U H G L n terms of stimulating teamwork. Hence, high incentives will most likely go together with more intense use of lateral coordination. Furthermore, the high correlation between trust and the knowledge-sharing variables was noticed. The correlations between the three dummies for the three categories of public sector organizations revealed that espe cially government institutions correlated with several of the independent variables. State This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 592 Journal of Public Administration Research and Theory enterprises were not correlated to any of the other variables. Only the dummy variable government institutions was included in the multivariate analyses because analysis of variance indicated that only government institutions were significantly differing from the other two categories of public sector organizations for the variables lateral coordina tion, incentives, and identification (F = 3.236, F = 4.342, and F = 4.755, respectivelyyf . The proposed relationships between the variables were analyzed through structural equation modeling and linear multiple regression. Structural equation modeling allows us to analyze direct and indirect relationships between the dependent and independent vari ables and among the dependent variables at the same time. Hence, it allows the testing and exploration of complex and new relationships, which is especially appropriate when indi cations in the literature on the relationships between the variables are still rather weak. In short, structural equation modeling can add additional insights to the more traditional regression analysis (Kline and Klammer 2001yf + R Z H Y H U L Q W H U D F W L R Q H I I H F W V D U H V W L O O E H V t analyzed with linear regression analysis, thus used here to test the interaction effect in two propositions (Id and 3byf . We built our structural equation model using Amos (Byrne 2001yf 7 K H R Y H U D O O P R G H l fit was tested using several fit indices, all indicating a good fit. Hence, conclusions about the relationships in the model may be drawn. There was a nonsignificant/? value (.21yf I R r the chi-square test indicating a good fit. The root mean square residual was low, .027; adjusted goodness of fit was .961 ; comparative fit index was .993; and the normed fit index was .969, all well above the threshold (.9yf J H Q H U D O O \ F R Q V L G H U H G Q H F H V V D U \ I R U D V D W L V I D F W R U y model fit (Kline 1998yf 7 K H ? G I \f value referred to a good fit as well, with 1.250. Structural equation modeling is based on the same conditions of normality, such as line arity, homoscedasticity, heteroscedasticity, and independence of error terms, required in many multivariate analyses. Although multiple responses per organization can jeopardize these conditions, none of the conditions were violated in our sample. Table 1 lists the structural parameter estimates of our path model. Structural equation parameter estimates are similar to standardized beta coefficients in regression analysis, and the critical ratio in dicates the significance of the relationship. Figure 1 displays the same information as table 1 in a more schematic way. Both are only displaying the statistical significant relationships. The model provided evidence for our propositions and also revealed new relation ships. The variable incentives, included here as a control variable, had a significant effect on knowledge sharing and explained part of the variance in the knowledge-sharing varia bles. However, our coordination variables were very relevant as well and clearly influenced the knowledge-sharing variables. The effect of formal systems on the intensity of knowl edge sharing was negative, confirming proposition la; however, there was no significant negative effect on knowledge-sharing effectiveness. The data also provided support for our second proposition (lbyf / D W H U D O F R R U G L Q D W L R Q U H V X O W H G L Q K L J K H U N Q R Z O H G J H V K D U L Q J L Q W H n sity and higher effectiveness of knowledge sharing. However, although the use of informal coordination did not result in higher knowledge-sharing intensity, the knowledge shared through informal interunit coordination was effective, partly confirming proposition lc. Hence, more use of horizontal, networklike, coordination had a positive qualitative and quantitative effect on knowledge sharing. There was a dominant positive effect of trust, strongly confirming proposition 2a. However, proposition 2b on the effect of identification could not be fully supported. Although identification did not result in higher knowledge sharing intensity, high levels of identification gave more effective knowledge sharing. The effect of power games was not significant, providing no support for proposition 3a. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 593 Table 1 Structural Parameter Estimates and Goodness-of-Fit Indices for the Path Model Paths Estimates Critical Ratio Proposition la: effect of formal systems Formal systems ===> knowledge-sharing intensity ?0.127 ?2.918 Proposition lb: effect of lateral coordination Lateral coordination ===> knowledge-sharing effectiveness 0.157 2.873 Lateral coordination ===> knowledge-sharing intensity 0.182 3.582 Proposition lc: effect of informal coordination Informal coordination ===> knowledge-sharing effectiveness 0.135 2.973 Proposition Id: relationship between lateral and informal coordination Lateral coordination ===> informal coordination 0.197 3.757 Proposition 2a: effect of trust Trust ===> knowledge-sharing intensity 0.236 5.241 Trust ===> knowledge-sharing effectiveness 0.328 7.006 Proposition 2b: effect of identification Identification ===> knowledge-sharing effectiveness 0.121 2.540 Proposition 2c: effect of government institutions Government institutions ===> identification ?0.120 ?2.336 Proposition 3a-3c: effect of power games Power games ===> informal coordination 0.294 5.623 Formal systems ===> power games 0.264 5.030 Lateral coordination ===> power games ?0.240 ?4.428 Identification ===> power games ?0.199 ?3.793 Power games ===> trust -0.400 -7.881 Effect of control variable incentives Incentives ===> knowledge-sharing effectiveness 0.187 3.422 Incentives ===> knowledge-sharing intensity 0.144 2.824 Government institutions ===> incentives ?0.097 ?2.144 Lateral coordination ===> incentives 0.499 10.581 Formal systems ===> incentives 0.115 2.450 Incentives ===> identification 0.162 2.665 Incentives ===> trust 0.141 2.721 Relationships between the independent variables Lateral coordination ===> formal systems 0.297 5.758 Lateral coordination ===> identification 0.195 3.240 Formal systems ===> trust 0.109 2.123 Relationship between the two dependent variables Knowledge-sharing effectiveness ===> knowledge-sharing intensity 0.355 7.311 Effect of government institutions on other variables (not included in the propositionsyf Government institutions ===> lateral coordination ?0.109 ?2.021 Government institutions ===> informal coordination 0.128 2.491 Note: normed fit index = .969, root mean square residual = .027, comparative fit index = .993, adjusted goodness of fit = .961, ?2(dfyf U Y D O X H . This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 594 Journal of Public Administration Research and Theory Figure 1 Path Model for Knowledge Sharing Government institutions Formal systems 0.30 Lateral Knowledge sharing effectiveness 0.36 Knowledge sharing intensity 0.24 ? dotted lines indicate negative relationships However, there were several relationships between power and other independent variables, resulting in an indirect relationship between power and knowledge sharing. The model revealed that power games and informal coordination went hand in hand. Although power games themselves had no direct effect on the knowledge sharing varia bles, they had an indirect negative effect by reducing trust. It was remarkable that formal systems resulted not only in more trust but also in more power games. Power games were influenced by two other variables as well. We found more power games in situations where there were fewer lateral relationships and lower identification. Trust was created in sit uations with many incentives for cooperation, few power games, and the presence of formal systems. People had higher levels of identification when there were more incentives given for cooperation and more formal lateral relationships. Lateral coordination seemed to imply the use of formal systems as well. The path model indicated that government institutions had different levels of the knowledge-sharing variables because of the lower use of lateral coordination, fewer incentives, lower levels of identification, and more use of informal networks. This created a context in government institutions that was less favorable for knowledge sharing. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 595 Proposition 2c on the lower levels of identification in public sector organizations was thus confirmed. However, there was no evidence that government institutions, often ac cused of red tape and use of several bureaucratic systems, made more use of formal systems as coordination mechanisms compared to the other two types of public sector organizations. The structural equation model also did not support the assumption that government institutions had more power games. Hence, proposition 3b could not be supported. The structural equation model revealed that more lateral coordination resulted in the development of more informal coordination, which strengthened the positive effects of lateral coordination on knowledge-sharing effectiveness. This could have indicated that lateral and informal coordination complemented each other in their effect on the knowledge-sharing variables. However, testing the interaction effect between lateral coordination and informal coordination on knowledge-sharing intensity and knowledge sharing effectiveness provided no proof for proposition Id. The interaction effect of in formal and lateral coordination did not improve the regression models; that is, there was no significant change in i?2 when the interaction term was added to the regression models with, respectively, knowledge-sharing intensity and knowledge-sharing effectiveness as dependent variables. The following values were found in the regression model for re gression on knowledge-sharing intensity: Ai?2 = .001, ? = .028, and t value = .626; and in the regression model for regression on knowledge-sharing effectiveness: Ai?2 = .002, ? = -.041, and t value = -.882. The finding for proposition 3 c was unexpected. We expected that power games, in combination with informal coordination, have a negative effect on knowledge sharing; in other words, that informal coordination is especially beneficial for knowledge sharing when power games are absent. However, the structural equation model indicated that power games went together with more informal coordination, which in turn resulted in more knowledge-sharing effectiveness. Through this path, power games had a positive effect on knowledge sharing. Testing an interaction effect between power games and informal coordination using regression analyses gave a significant change in i?2 in the two regression models. We found the following values for regression on knowledge sharing intensity: Ai?2 = .011, ? = .109, and t value = 2.429; and for regression on knowledge-sharing effectiveness: Ai?2 = .013, ? = .119, and rvalue = 2.577. The inter action effect of informal coordination with power games also indicated that the presence of power games in the informal coordination had a significant positive effect on knowledge sharing intensity and on knowledge-sharing effectiveness. Thus, both models revealed a remarkable positive combination of power games and informal coordination. Finally, taking direct and indirect effects together, more intensive knowledge sharing occurred especially in the context of many lateral relationships, when supported by incen tives and when trust existed. Knowledge sharing was effective in the same situations. Effective knowledge sharing caused higher knowledge-sharing intensity. DISCUSSION Public sector organizations are often presented as strongly dominated by procedures and high formalization. Behn (1995yf I R U L Q V W D Q F H P H Q W L R Q H G W K D W D Y R L G L Q J D Q R Y H U O R D G R f procedural rules is one of the concerns of public managers and researchers in public This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 596 Journal of Public Administration Research and Theory management. Furthermore, according to some authors, motivation and commitment are lower in public sector organizations, another major concern for public sector managers (Behn 1995; Moon 2000yf 7 K L V L V Q R W W K H L G H D O H Q Y L U R Q P H Q W I R U N Q R Z O H G J H V K D U L Q J 9 D n den Bosch, Volberda, and de Boer 1999yf . However, evidence on the dominance of formal systems, with its many disadvantages (cf. the red tape argumentsyf L Q S X E O L F V H F W R U R U J D Q L ] D W L R Q V L V Z H D N % R \ Q H \f. Formal systems (mean = 3.16yf Z H U H D F F R U G L Q J W R R X U U H V S R Q G H Q W V V R P H Z K D W O H V V L P S R U W D Q W W K D n the use of lateral and informal coordinations (mean = 3.27 and 3.59, respectivelyyf + H Q F H , a dominance of systems in the cooperative episode was not observed. However, when the coordination of the cooperative episode was mainly based on formal systems, it resulted in lower knowledge-sharing intensity, although the impact was not very large (parameter estimate = ?0.127yf V H H W D E O H \f. We can therefore conclude that formal systems were not the main obstacle to knowledge sharing and that they were not even dominant in our broad sample of cooperative episodes in public sector organizations. In short, our sample did not reveal negative bureaucratic effects on knowledge sharing, even when we limited our sample to cooperative episodes in government institutions?the pure public organiza tions according to Fottler (1981yf . Lateral coordination was very important for the intensity and effectiveness of knowl edge sharing (parameter estimate = 0.157 and 0.182, respectivelyyf / D W H U D O F R R U G L Q D W L R n not only had a significant positive impact on intensity and effectiveness of knowledge sharing but also influenced many other variables, resulting in a strong indirect effect of lateral coordination on the knowledge-sharing variables. Although lateral coordination was necessary for knowledge sharing, it also had a negative effect via formal systems, decreasing knowledge sharing. Apparently, formal systems were necessary to support the working of lateral coordination. Informal coordination did not lead to higher knowledge-sharing intensity, although it did lead to more effective knowledge sharing. Literature on informal coordination has drawn our attention to the inflexibility of informal coordination and, hence, its limitations on the kind of knowledge that is shared and the parties involved in the knowledge sharing (Hansen 1999, 2002yf : H H [ S H F W H G W R I L Q G D S R V L W L Y H H I I H F W R Q W K H N Q R Z O H G J H V K D U L Q g variables, especially on the "intensity" of knowledge sharing. A less positive, smaller or even negative effect on the variable effectiveness of knowledge sharing was to be expected. On the contrary, the effectiveness of knowledge sharing was clearly positively influenced by the use of informal coordination. One explanation suggested by the literature is that informal coordination develops trust and openness toward accepting and thus applying the knowledge of others (Adler 2001; Hansen 2002yf + R Z H Y H U W K L V Z D V Q R t the case for our data because informal coordination scored low on trust (correlation = ?.12yf D Q G K L J K R Q S R Z H U J D P H V F R U U H O D W L R Q \f. Kostava and Kendall (2003yf U H I H U U H d to people in informal networks as a kind of boundary spanners connecting units. They said that although social networks are primarily established to achieve personal benefits, social networks can become public when a group of people or a unit in an organization can tap into the resources made available through the network. In fact, people able to share knowledge, able to see opportunities, and believing in the benefits of networking engage in networks (Burt 1992yf + H Q F H D O W K R X J K S H R S O H D U H F U H D W L Q J L Q I R U P D O F R R U G L Q D W L R Q Z L W h other departments to obtain knowledge for their own benefit, they help their department through this networking. Those people might be better able to find, share, and successfully apply the knowledge they need for their tasks. This could be an explanation for the positive This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 597 relationship between informal coordination and effectiveness of knowledge sharing in our data. It is interesting that in the cooperative episodes in the governmental institutions more informal coordination and less lateral coordination were used than was the case in the cooperative episodes in the other public sector organizations. The higher use of informal coordination could not compensate for the shortcomings in the lateral coordination because of the lower impact of informal coordination on knowledge-sharing intensity. Furthermore, the correlation between lateral and informal coordination suggests that lateral coordina tion was a source for informal coordination because it created opportunities for people to meet each other and develop personal relationships that could result in more informal networks as well in the cooperative episodes. Our study confirmed the dominant positive impact of trust on the knowledge-sharing variables, which has been recognized in several other studies. It was trust that was impor tant, not as part of informal or lateral coordination or developed through identification. Not only would people share more knowledge in an environment of trust but also knowledge sharing would be more effective. Power games and trust correlated strongly negative, as expected. Although power games had unexpectedly no direct effect on the knowledge sharing variables, it had a strong indirect effect through trust. Hence, power games and trust could be seen as two polar variables that characterize the organizational climate with more power games and low trust at one end and fewer power games and high trust at the other end, with less and more knowledge sharing, respectively. Another remarkable finding was that a combination of informal coordination and power games was leveraging intensity and effectiveness of knowledge sharing, strongly contradicting our assumption. This was even more surprising when taking into account the arguments of O'Toole and Meier (2004yf Z K R Z D U Q H G X V R I W K H J U H D W H U U L V N R I S R Z H U J D P H s in informal coordination in public sector organizations. Their arguments were not sup ported at all. Power games and informal coordination were highly correlated (.3yf 3 H R S O e involved in informal coordination were more knowledgeable than other people in the organization (Krackhardt 1990yf 7 K H \ N Q H Z Z K R W R F R Q W D F W D Q G Z K R K D G Z K D W N Q R Z O H G J H , and they understood the informal structure of the organization. This information gave these people many opportunities to share knowledge and, unfortunately, also for power games (Adler and Kwon 2002; Krackhardt 1990; Leana and Van Buren 1999yf + R Z H Y H U W K L V Z D s not yet a full explanation for the interaction effect found. Another explanation could be that in the presence of power games, people were only willing to share their knowledge if they felt protected against opportunistic behavior. Informal networks can provide a sense of protection and thus induce people to share their knowledge. Hence, the presence of power games could have resulted in the development of informal coordination in the cooperative episode. It might be that people engaged in or built informal coordination especially to cope with power games (Gresov and Stephans 1993yf $ O W K R X J K I X U W K H U V W X G \ L V Q H F H V V D U y to unpack this unexpected role of power games, there was evidence pointing to the use of informal coordination to cope with power games. However, this was not especially the case in government institutions. There was evidence on the impact of low organizational commitment in government institutions. The level of identification was lower in pure public organizations, the gov ernment institutions, and lower identification impacted on the effectiveness of knowledge sharing?see evidence for proposition 2b and 2c in table 1. Hence, government institu tions might risk facing knowledge-sharing problems due to lower levels of identity. This conclusion confirmed the assumptions in the public sector literature on commitment This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 598 Journal of Public Administration Research and Theory (Boyne 2002; Moon 2000yf , Q G H H G L Q W K H J R Y H U Q P H Q W D O L Q V W L W X W L R Q V L Q R X U V D P S O H W K H H Q d product was less definable, one of the reasons for less commitment and identification. However, a much larger danger rested in the fewer incentives to stimulate cooperative behavior in government institutions. Although the impact of government institutions ver sus other public sector organizations on incentives was not very strong (parameter esti mate = ?0.097yf V H H W D E O H \f, the combination of low identity and low incentives could result in a serious lower level of knowledge-sharing effectiveness. Incentives based on intrinsic and extrinsic rewards have been indicated in the literature as very important for the level of organizational commitment (including identityyf D P R Q J S X E O L F V H F W R U H m ployees (Khojasteh 1993; Young, Worchel, and Woehr 1998yf 7 K L V Z D V F R Q I L U P H G L n our study. Respondents reported lower identification, together with few incentives, for cooperative behavior and less lateral coordination. Values of the three correlated variables were all lower in the cooperative episodes in government institutions. Cooperative behav ior and knowledge sharing were not yet recognized as important, or at least, they were not much formally encouraged in the organizations in our sample. The incentives had a positive impact on identity and also on trust. Hence, through the direct and indirect effects, knowledge-sharing effectiveness could be greatly influenced by more incentives. CONTRIBUTIONS AND LIMITATIONS We have contributed to the public sector literature by indicating which organizational characteristics are important and might need to be altered to make knowledge management enter the (knowledge intensiveyf S X E O L F V H F W R U : H K D Y H D O V R F R Q W U L E X W H G W R S X E O L F P D n agement literature by revisiting the organization design fundamentals, namely, coordina tion choices, and assessing their relevance in the knowledge-sharing discussion in the public sector. Our study was focused on the sharing of know-how between departments during co operative episodes. We learned that the most appropriate organization design consists of coordination that is not solely based on formal systems but also more on lateral coordination, combined with high levels of trust and identification, clear incentives, and the absence of power games. This confirmed most of our propositions except for some on power games and informal coordination. Applying knowledge management principles in public organizations requires special attention to power games. Furthermore, cooperative behavior organized through lateral coordination should be formally evaluated, rewarded, and stimulated; in other words, incentives for cooperative behavior are required. Stimulating informal co ordination was, according to our data, of little use except in organizations where power games are widespread. However, in these organizations, it is preferable to try to reduce power games instead of trying to reach knowledge sharing by informal coordination. Trust was undoubtedly of major importance, and any measure increasing trust will pay off through better and more knowledge sharing. The unexpected findings on informal coordination and power games have contributed to the network literature, which paid quite a lot of at tention to the knowledge-sharing issue, but which, in our opinion, needs to refine its studies more. We acknowledge that our study has its limitations. Data collection occurred through a cross-sectional survey. This meant an inability to look for longitudinal effects or to study processes. The concept of knowledge and knowledge sharing is hard to operationalize and This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 599 to measure quantitatively in larger samples. Hence, we recognize that our measurements are only proxies for the complex process of knowledge sharing. The sample of our study consisted of cooperative episodes in Belgian public sector organizations. Hence, the findings are only representative of the Belgian popula tion of public sector organizations. The absolute level of the variables measured, such as power games, identity, formal systems, and incentives, might differ in various coun tries. Afonso, Schuknecht, and Tanzi (2005yf V K R Z H G W K D W % H O J L X P V F R U H V U H O D W L Y H O \ K L J h in terms of bureaucracy measured by red tape compared to other European countries and the United States. Consequently, more formal systems might be found in Belgium. However, there is no indication that the relationships found would be very typical for Belgium. Apart from validity concerns, our study was limited in scope. Only knowledge sharing during cooperative episodes and only current sharing of know-how and its effectiveness were studied, although knowledge sharing often has invisible and/or long-term effects. Future research could address this issue. Qualitative data could explain more in depth what is occurring in knowledge-sharing processes. Knowledge-sharing events are complex pro cesses that can be influenced by a large number of variables. Our study is also limited in scope in the number of variables included. An interesting extension of our research, for instance, would be to include complexity of the cooperative episodes in interaction with the coordination mechanisms. APPENDIX 1 Questionnaire items Power: based on Devos, Vanderheyden, and Van den Broeck (2001yf In meetings, it is always the same people, and seldom the most competent ones, who talk too much. Within our organization, power games between the units play an important role. Relationships between people of different units can be rather strained. In our organization, favoritism is an important way to achieve something. Staff members are sometimes taken advantage of in our organization. Identification: based on Wagner and Moch (1986yf People in my work group are willing to make sacrifices for the sake of the work group (such as working late now and then, going out of their way to helpyf . People in my work group do realize that they sometimes are going to have to make sacrifices for the sake of the work group as a whole. People are aware that if they are going to be a part of the work group, they are sometimes going to have to do things they do not want to do. People in my work group do their best to cooperate with each other instead of trying to work things out on their own. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 600 Journal of Public Administration Research and Theory Trust: based on McAllister (1995yf The people in the other unit approach their job with professionalism and dedication. Given the other unit's track record, I see no reason to doubt their competence and preparation for the job. I can rely on the people from the other unit not to make my job more difficult by careless work. Most people, even those who are not close friends to people in the other unit consider the people of the other unit to be trustworthy. Lateral coordination: based on Miller and Droge (1986yf I L U V W I R X U L W H P V \f and completed with two new items Interunit committees are set up to allow units to engage in joint decision making. Task forces (project teamsyf D U H V H W X S W R I D F L O L W D W H L Q W H U X Q L W F R O O D E R U D W L R Q R Q D V S H F L I L c project. There are people with a coordinating role whose specific job it is to coordinate the efforts of several departments for purposes of a specific project. Decision making in our organization is characterized by participative, cross-functional dis cussions in which different departments, functions, or divisions get together. There are process responsible whose specific job it is to exchange knowledge and information for purposes of a specific process. Information and experiences are often shared in meetings or during teamwork. Formal systems: newly developed items Formal procedures determine how we work together with the other unit. Information is mainly held in and exchanged through a large number of reports and formal documents. In general, our work is subject to a large number of rules. The information that is required to do my job is laid down in procedures, goals and rules. Our direct supervisors decide how we should execute our tasks. Communication with other units occurs via the unit heads. Knowledge and information exchange between units is the responsibility of the unit heads. Informal coordination: newly developed items I have my informal network of friends in the organization. Having your own personal network is important in our organization. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 601 Personal networks are necessary to learn from others. Personal networks are necessary to do your job efficiently in our organization. I have my own network with people that can advice me when I face a problem or need help. Incentives: newly developed items We are evaluated as a team. Teamwork is included in our yearly evaluation. Cooperative behavior is explicitly rewarded in our organization (e.g., through bonuses, promotion, and formal evaluationsyf . Exchanging knowledge and information between units is formally encouraged. Cooperation between units is one of the strategic objectives of our organization. Interunit knowledge sharing is one of the priorities of our management. Knowledge-sharing intensity: newly developed items Knowledge is frequently shared between the two units. We frequently provide information to the other unit. We share our expertise and knowledge with the other unit. There is frequent exchange of ideas and experiences between the two units. We share knowledge about our way of working with the other unit. There is a good exchange of knowledge between the units. Effectiveness of knowledge sharing: based on Wathne, Roos, and von Krogh (1996yf I L U V t three itemsyf D Q G % H F H U U D ) H U Q D Q G H ] D Q G 6 D E K H U Z D O \f (last two itemsyf To what extent did you acquired knowledge that generally caused you to develop new insights. To what extent did you receive knowledge that enabled you to see new ways of performing current tasks within your organization. To what extent did the cooperative projects with the other units enable you to perform new tasks as a result of acquired knowledge. The available knowledge improves your effectiveness in performing your tasks. The available knowledge improves my units overall effectiveness. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms O? o APPENDIX 2 Table AI Means, Standard Deviations, Intercorrelations, and Cronbach's Alpha values (on the Diagonalyf R I W K H 9 D U L D E O H V 0 H D V X U H d Variable Mean Standard deviation 1 1. Formal systems 3.16 0.63 2. Lateral coordination 3.27 0.66 3. Informal coordination 3.59 0.63 4. Power games 3.02 0.77 5. Trust 3.51 0.68 6. Identification 3.81 0.53 7. Incentives 2.90 0.64 8. Knowledge-sharing intensity 3.41 0.76 9. Effectiveness of knowledge sharing 3.44 0.61 10. Government institution ? ? 11. Public sector institutions ? ? 12. State enterprises ? ? .71 .31** .06 .19** .08 .15** .25** .01 .05 .01 .04 .04 .71 .12* -.22** .20** .29** .54** .41** .37** .12* .14** .03 .76 .30** .12* .02 .01 .04 .12* .12* .10 .02 .79 .40** -.22** .20** -.32** ,22** .05 .08 .05 84 18** 25** 45** 43** 07 12 06 77 28** 26** 27** 15** 10 07 .69 .41** .39** .15** .13* .03 .90 .57** .07 .09 .02 .82 -.05 .06 -.01 *p< .05, **p< .01. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 603 REFERENCES Adler, Paul S. 2001. Market, hierarchy, and trust: The knowledge economy and the future of capitalism. Organization Science 12:215?34. Adler, Paul S., and Seok-Woo Kwon. 2002. Social capital: Prospects for a new concept. Academy of Management Review 27:17-40. Afonso, Antonio, Ludger Schuknecht, and Vito Tanzi. 2005. Public sector efficiency: An international comparison. Public Choice 123:321^47. Alvesson, M. 1993. Organizations as rhetoric: Knowledge-intensive firms and the struggle with ambiguity. Journal of Management Studies 30:997?1015. -. 2000. Social identity and the problem of loyalty in knowledge-intensive companies. Journal of Management Studies 37:1107?23. -. 2001. Knowledge work: Ambiguity, image and identity. Human Relations 54:863?86. Andrews, Kate, and Brian L. Delahaye. 2000. Influences on knowledge processes in organizational learning: The psychosocial filter. Journal of Management Studies 73:797?810. Argote, L. 1982. Input uncertainty and organizational coordination in hospital emergency units. Administrative Science Quarterly 27:420?34. Argote, L., P. Ingram, J. M. Levine, and R. L. Moreland. 2000. Knowledge transfer in organizations: Learning from the experience of others. Organizational Behavior and Human Decision Processes 82:1-8. Argote, L., Bill McEvily, and R. Reagans. 2003. Introduction to the special issue on managing knowledge in organizations: Creating, retaining and transferring knowledge. Management Science 49 (4yf . Ashforth, Blake E., and Fred Mael. 1989. Social identity theory and the organization. Academy of Management Review 14:20-39. Ayas, Karen, and Nick Zeniuk. 2001. Project-based learning: Building communities of reflective practitioners. Management Learning 32:61-76. Bate, S. P., and G. Robert. 2002. Knowledge management and communities of practice in the private sector: Lessons for modernizing the National Health Service in England and Wales. Public Administration 80:643?33. Becerra-Fernandez, Irma, and Rajiv Sabherwal. 2001. Organizational knowledge management: A contingency perspective. Journal of Management Information Systems 18 (1yf " . Behn, Robert D. 1995. The big questions of public management. Public Administration Review 55:313?24. Benn, S. I., and G. F. Gaus. 1983. Public and private in social life. New York: St. Martin's Press. Berman Brown, Reva, and Martyn J. Woodland. 1999. Managing knowledge wisely: A case study in organisational behaviour. Journal of Applied Management Studies 8:175?98. Birkinshaw, Julian, Robert Nobel, and Jonas Ridderstrale. 2002. Knowledge as a contingency variable: Do the characteristics of knowledge predict organization structure? Organization Science 13:274-89. Boyne, George A. 2002. Public and private management: What's the difference? Journal of Management Studies 39:97-122. Bozeman, B. 1987. All organizations are public. London: Jossey-Bass. Burt, Ronald S. 1992. Structural holes: The social structure of competition. Cambridge, MA: Harvard Univ. Press. Byrne, Barbara M. 2001. Structural equation modeling with Amos: Basic concepts, applications and programming. Mahwah, NJ: Lawrence Erlbaum associates. Cabrera, Angel, and Elisabeth Cabrera. 2002. Knowledge-sharing dilemmas. Organization Studies 23:687-710. Ciborra, Claudio U., and Rafael Andreu. 2001. Sharing knowledge across boundaries. Journal of Information Technology 16:73?81. Constant, D., S. Kiesler, and L. Sproull. 1994. What's mine is ours, or is it? A study of attitudes about information sharing. Information Systems Research 5:400-21. De Cremer, David, Mark Snyder, and Siegfried Dewitte. 2001. 'The less I trust, the less I contribute (or notyf " 7 K H H I I H F W V R I W U X V W D F F R X Q W D E L O L W \ D Q G V H O I P R Q L W R U L Q J L Q V R F L D O G L O H P P D V ( X U R S H D n Journal of Social Psychology 31:93?107. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 604 Journal of Public Administration Research and Theory Devos, Geert, Karlien Vanderheyden, and H. Van den Broeck. 2001. A framework for assessing commitment to change: Process and context variables of organizational change. AOM, Washington, August 3?8. Duncan, R. B., and A. Weiss. 1979. Organizational learning: Implications for organizational design. Research in Organizational Behavior 1:75?123. Easterby-Smith, Mark, and Marjorie Lyles. 2003. Introduction: Watersheds of organizational learning and knowledge management. In Handbook of organizational learning and knowledge management, ed. Mark Easterby-Smith and Marjorie Lyles, 1?15. Oxford: Blackwell Publishing. Fottler, Myron D. 1981. Is management really generic? Academy of Management Review 6:1?12. Galbraith, J. R. 1973. Designing complex organizations. Reading, MA: Addison-Wesley Publishing Company. Gargiulo, Martin, and Mario Benassi. 2000. Trapped in your own net? Network cohesion, structural holes, and the adaptation of social capital. Organization Science 11:183?96. Ghoshal, Sumantra, Harry Korine, and Gabriel Szulanski. 1994. Interunit communication in multinational corporations. Management Science 40 (1yf . Goulet, Laurel R., and Margaret L. Frank. 2002. Organizational commitment across three sectors: Public, non-profit, and for-profit. Public Personnel Management 31:201?10. Government. 2004. Statistics of the federal personnel. Provided by PDATA, Belgian Federal Government Service for Economy, SME, and Energy. Grandori, A. 1997. Governance structures, coordination mechanisms and cognitive models. The Journal of Management and Governance 1:29?47. Grant, R. M. 1996. Toward a knowledge-based theory of the firm. Strategic Management Journal 17 (Winter Specialyf . Gresov, C, and Carroll Stephans. 1993. The context of interunit influence attempts. Administrative Science Quarterly 38:252-70. Grover, Varun, and Thomas H. Davenport. 2001. General perspectives on knowledge management: Fostering a research agenda. Journal of Management Information Systems 18 (1yf " . Hansen, M. T. 1999. The search-transfer problem: The role of weak ties in sharing knowledge across organization subunits. Administrative Science Quarterly 44:82?111. -. 2002. Knowledge networks: Explaining effective knowledge sharing in multiunit companies. Organization Science 13:232-^4-8. Harmon, M., and R. Mayer. 1986. Organization theory for public administration. Boston, MA: Little Brown. Haynes, Philip. 2005. New development: The d?mystification of knowledge management for public services. Public Money & Management 25:131?5. Heffron, Florence. 1989. Organization theory and public organizations: The political connection. Englewood Cliffs, NJ: Prentice Hall. Hood, C. 1991. A public management for all seasons? Public Administration 69:3?19. Hood, C, and G. Peters. 2004. The middle aging of New Public Management: Into the age of paradox? Journal of Public Administration Research and Theory 14:267?82. Husted, Kenneth, and Snejina Michailova. 2002. Diagnosing and fighting knowledge-sharing hostility. Organizational Dynamics 31:60-73. Jarvenpaa, Sirkka L., and D. Sandy Staples. 2001. Exploring perceptions of organizational ownership of information and expertise. Journal of Management Information Systems 18 (1yf " . K?ser, Philipp A. W., and R. E. Miles. 2002. Understanding knowledge activists' successes and failures? Long Range Planning 35:9-28. Khojasteh, Mak. 1993. Motivating the private vs public sector managers. Public Personnel Management 22:391-^01. Kline, Rex B. 1998. Principles and practices of structural equation modeling. New York: The Guilford. Kline, Theresa J. B., and Joy D. Klammer. 2001. Path model analyzed with ordinary least squares multiple regression versus Lisrel. Journal of Psychology 135:213?25. Kogut, B., and U. Zander. 1996. What firms do? Coordination, identity, and learning. Organization Science 7:502-18. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms Willem and Buelens Knowledge Sharing in Public Sector Organizations 605 Kostava, Tatiana, and Kendall Roth. 2003. Social capital in multinational corporations and a micro macro model of its formation. Academy of Management Review 28:297?318. Krackhardt, David. 1990. Assessing the political landscape: Structure, cognition, and power in organizations. Administrative Science Quarterly 35:342?69. Leana, Carrie R., and Harry J. Van Buren. 1999. Organizational social capital and employment practices. Academy of Management Review 24:538?55. Luen, Tan Woei, and Suliman Al-Hawamdeh. 2001. Knowledge management in the public sector: Principles and practices in policy work. Journal of Information Science 27:311?8. Maguire, Steve, Nelson Phillips, and Cynthia Hardy. 2001. When 'silence=death', keep talking: Trust, control and the discursive construction of identity in the Canadian HIV/AIDS treatment domain. Organization Studies 22:285?310. March, R. M., and H. A. Simon. 1958. Organizations. New York: Wiley International. Martinez, Jon I., and J. Carlos Jarillo. 1989. The evolution of research on coordination mechanisms in multinational corporations. Journal of International Business Studies 20:489?514. McAllister, D. J. 1995. Affect-and cognition-based trust as foundations for interpersonal cooperation in organizations. Academy of Management Journal 38:24?59. McEvily, Bill, Vincenzo Perrone, and Akbar Zaheer. 2003. Trust as an organizing principle. Organization Science 14:91?103. Miller, D., and C. Droge. 1986. Psychological and traditional determinants of structure. Administrative Science Quarterly 31:539-60. Mintzberg, H. 1989. Mintzberg on management: Inside our strange world of organizations. 1st ed. New York: The Free Press. Moe, C, and R. Gilmour. 1995. Rediscovering principles of public administration: The neglected foundation of public law. Public Administration Review 54:111?22. Moon, M. Jae. 2000. Organizational commitment revisited in new public management: Motivation, organizational culture, sector, and managerial level. Public Performance & Management Review 24:177-94. Newell, Sue, and Jacky Swan. 2000. Trust and inter-organizational networking. Human Relations 53:1287?327. Nidumolu, S. R. 1996. A comparison of the structural contingency and risk-based perspectives on coordination in software-development projects. Journal of Information Systems 13:77?113. NOSS. 2005. Statistics on 'Employment in the fourth quarter of 2004.' National Office for Social Security. Nunnally, J. C. 1978. Psychometric theory. New York: McGraw-Hill. Olsen, Johan P. 2006. Maybe it is time to rediscover bureaucracy. Journal of Public Administration Research and Theory 16:1?24. Osborne, David, and Ted Gaebler. 1993. Reinventing government: How the entrepreneurial spirit is transforming the public sector. New York: Penguin Books. O'Toole, Laurence J., and Kenneth J. Meier. 2004. Desperately seeking Selznick: Cooptation and the dark side of public management in networks. Public Administration Review 64:681?93. Perry, James L., and L. Porter. 1982. Factors affecting the content for motivation in public organizations. Academy of Management Review 7:89?98. Perry, James L., and Hal G. Rainey. 1988. The public-private distinction in organization theory: A critique and research strategy. Academy of Management Review 13:182?201. Podsakoff, Philip M., and Dennis W. Organ. 1986. Self-reports on organizational research: Problems and prospects. Journal of Management 12:531-^4. Posner, Barry Z., and Warren H. Schmidt. 1996. The values of business and federal government executives: More different than alike. Public Personnel Management 25:277?89. Rainey, Hal G. 1997. Understanding and managing public organizations. San Francisco, CA: Jossey Bass Publishers. Rainey, Hal G., and Barry Bozeman. 2000. Comparing public and private organizations: Empirical research and the power of the a priori. Journal of Public Administration Research and Theory 10:447-69. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms 606 Journal of Public Administration Research and Theory Robertson, Maxime, Harry Scarbrough, and Jacky Swan. 2003. Knowledge creation in professional service firms: Institutional effects. Organization Studies 24:831?57. Robertson, Maxime, and Jacky Swan. 2003. 'Control-what control?' Culture and ambiguity within a knowledge intensive firm. The Journal of Management Studies 40:831?58. Sanchez, R., and A. Heene, eds. 1997. A competence perspective on strategic learning and knowledge management. Strategic learning and knowledge management, 3?18. New York: John Wiley & Sons. Scarbrough, Harry. 2003. Why your employees don't share what they know. Knowledge Management Review 6 (3yf . Senge, P. M. 1990. The fifth discipline. London: Century Business. Spector, Paul E. 1994. Using self-report questionnaires in OB research: A comment on the use of a controversial method. Journal of Organizational Behaviour 15:385?92. -. 2006. Method variance in organizational research. Organizational Research Methods 9:221?32. Starbuck, W. H. 1992. Learning by knowledge-intensive firms. Journal of Management Studies 29:713? 740. Szulanski, G. 2000. The process of knowledge transfer: A diachronic analysis of stickiness. Organizational Behavior and Human Decision Processes 82:9-27. Teece, David F. 1998. Capturing value from knowledge assets: The new economy, markets for know how, and intangible assets. California Management Review 40 (3yf . Tsai, Wenpin. 2002. Social structure of "coopetition" within a multiunit organization: Coordination, competition, and intraorganizational knowledge sharing. Organization Science 13:179-90. Van De Ven, A. H., A. L. Delbecq, and R. Koenig. 1976. Determinants of coordination modes whitin organizations. American Sociological Review 41 (Aprilyf " . Van den Bosch, F. A. J., H. W. Volberda, and M. de Boer. 1999. Coevolution of firm absorptive capacity and knowledge environment: Organizational forms and combinative capabilities. Organization Science 10:551-68. Wagner, John A., and Michael K. Moch. 1986. Individualism-collectivism: Concept and measure. Group & Organization Studies 11:280-304. Wathne, K., J. Roos, and G. von Krogh. 1996. Towards a theory of knowledge transfer in a cooperative context. In Managing knowledge: Perspectives on cooperation and competition, ed. von Krogh and J. Roos, 55?81. London: Sage Publications. Young, Brian S., S. Worchel, and David J. Woehr. 1998. Organizational commitment among public service employees. Public Personnel Management 27:339?48. This content downloaded from 69.67.123.93 on Mon, 03 Oct 2016 19:59:45 UTC All use subject to http://about.jstor.org/terms