• 沒有找到結果。

資訊教師知識分享與其潛在社會網路

N/A
N/A
Protected

Academic year: 2021

Share "資訊教師知識分享與其潛在社會網路"

Copied!
133
0
0

加載中.... (立即查看全文)

全文

(1)國立臺灣師範大學資訊教育研究所 博 士 論文 指導教授: 邱貴發 博士. 資訊教師知識分享與其潛在社會網路 Knowledge Sharing of ICT Teachers and its Underlying Social Networks. 研究生: 林芳苓 撰 中華民國 九十八 年 十二 月.

(2) i.

(3) 摘 要 本論文研究知識分享背後之社會現象,包含兩個研究階段。在第一個研究 階段中,發展一個包含知識分享論壇的線上實務社群。 此階段的目標在建立支 援專業發展且由參與者自主的線上社群。第二個研究階段則探討知識分享者的 潛在社會網路,期能發現有利於促進知識分享的社會關係,找出資訊組長之間 知識分享的缺陷,及探討他們專業發展的需求。 研究以社會實踐理論為基礎,運用知識聲望的概念驗證個人社會網路之關 係屬性、位置優勢與知識轉介機會。研究使用社會網路分析之自我中心網路方 法進行訪談,調查及相關分析處理,應用二次式分派程序(quadratic assignment procedure, QAP)計算多元關係之間的相關性,使用徑路分析檢驗知識聲望與個 人知識分享關係屬性的相關性,及使用多元迴歸探討資訊組長在校際間的關係 屬性與知識聲望的相關性。 研究結果證明以知識聲望作為知識分享指標的可行性,同時也揭露出大部 分的資訊組長個人知識網路薄弱,他們在資訊科技實務分享網路與資訊融入教 學實務分享網路的對象並不相似。具有較高知識聲望的資訊組長貢獻知識給線 上社群比貢獻知識給其他學校的資訊組長機會高;因此他們較少發展幫助其他 學校資訊組長的網路;同時他們也只與幫助過他們的資訊組長發展較密切的關 係。線上知識貢獻這因素對於個人知識網路有限的資訊組長獲得知識聲望有中 介效果。知識貢獻網路中的結構洞也有促進知識聲望的效果。資訊融入教學種 子學校的資訊組長能將知識仲介給其他學校資訊組長。 此研究發展出使用聲望代替工作績效做為知識工作者的指標,以檢查知識 分享的特性及說明它可應用於實務社群的研究。本研究結果能說明資訊組長社 群中知識傳遞缺乏的現象。本研究的發現也可關連到其他領域的資訊科技實務。 關鍵字: 知識分享;知識管理;社會網路;實務社群;聲望;結構洞;仲介. i.

(4) i.

(5) Abstract This dissertation investigates the social phenomena behind knowledge sharing. It encompasses two research stages. In the first stage, an online community of practice that uses open discussion forums is developed. The objective is to create a self-sustaining online community that supports professional development. In the second stage, the social networks underlying knowledge sharing are explored with the objectives of uncovering facilitative relationships, ascertaining the deficiencies of knowledge sharing among school technology coordinators (STCs), and exploring the coordinators’ needs for professional development. A concept of knowledge prestige was proposed on the basis of the theory of social practices, and hypotheses were developed to verify the relational properties of personal networks, positional advantages, and brokerage opportunities. Social network analysis (SNA) was applied to conduct the egocentric network interview, surveys, and related analyses. The quadratic assignment procedure (QAP) was used to compute the correlations of multiplex relationships. Path analysis was applied to examine the association between knowledge prestige and the relational properties of personal knowledge sharing. Multiple regression equations were used to explore the association between relational properties across school boundaries and knowledge prestige. The results verified the usability of knowledge prestige as an indicator of knowledge sharing. The results also revealed that most STCs have weak personal knowledge networks. The network correlates of ICT practices and ICT-in-education practices are not similar. Prestigious STCs engage more in contributing knowledge to online communities than to STCs at other schools. They do not expend as much effort in helping other STCs and intimately reciprocate only to a smaller group from which i.

(6) they themselves have received advice. Online knowledge contributions have mediation effects for STCs who own limited personal networks, enabling them to acquire prestige. The structural hole of knowledge contribution affects knowledge prestige. STCs of ICT-in-education exemplar schools can act as information brokers to STCs of other schools. Several new ways of evaluating the performance of knowledge sharing have been developed during the present research. This research uses prestige, rather than job evaluations, as an indicator for knowledge-intensive workers, to examine the characteristics of knowledge sharing and demonstrate applicability to studies of a community of practices. The results reveal deficiencies in knowledge dissemination among STC communities. These findings hold relevance for studies on the information technology practices of other professionals.. Keywords: Knowledge Sharing, Knowledge Management, Social Networks, Community of Practices, Prestige, Structural Hole, Brokerage. ii.

(7) Acknowledgment I would particularly like to thank Dr. Guey-Fa Chiou, my advisor, who provided continuous support, encouragement, and advice, and yet made sure at every juncture that this dissertation research was in the correct direction and had academic merit. He is also very generous and has an excellent work ethic, and I hope that I can live up to his example in my future career. Gratitude is also extended to the professors of my dissertation committee for their knowledgeable and helpful suggestions throughout the process. I must thank the ICT teachers who agreed to participate in this research through survey responses and by sharing their experiences during interviews. This dissertation research clearly could not have happened without their contributions. I would like to thank Yu-Fang Kang and Li-Ting Cai for their cooperation in collecting and coordinating data. Without them, this dissertation could not have been completed. I also want to mention the students in the Network Learning Research Group of the Information and Computer Education Institute, who created a wonderful climate of mutual support and friendship that I think is one of the best assets of my Ph. D. program. I would also like to thank my parents for their love, for setting an apt example, and for their constant support. Fortunately, their optimism and thoughtful curiosity about the world have been passed down to me. Finally, I must thank my husband—Chun-Jen Lee—for his support and encouragement that kept my research moving and gave it meaning.. iii.

(8) iv.

(9) Table of Contents List of Tables........................................................................................................... i Lists of Figures ......................................................................................................ii 1. Introduction....................................................................................................... 1 1.1 Statement of Problem ...................................................................................... 1 1.1.1 Social Networks underlying Social Practices ........................................ 3 1.1.2 Social Resources of STCs’ Community................................................. 5 1.2 Research Purposes ........................................................................................... 7 1.3 Operational Definitions ................................................................................... 7 1.4 Structure of the Dissertation............................................................................ 8 2. Literature Review ........................................................................................... 11 2.1. Theoretical Foundations ............................................................................... 11 2.1.1. Network Benefits of Knowledge Sharing ........................................... 12 2.1.2. Gift-Giving Attitudes of Knowledge Sharing ..................................... 13 2.1.3. Positional Advantage of Knowledge Sharing ..................................... 15 2.2 Methodological Foundations ......................................................................... 16 2.2.1 Social Network Analysis...................................................................... 17 2.2.1.1 Network Content ........................................................................... 17 2.2.1.2. Ego-centric Network Analysis...................................................... 18 2.2.2. Social Resource................................................................................... 19 2.2.2.1. Relational Properties of PSNs ..................................................... 20 2.2.2.2. Centrality ..................................................................................... 21 2.2.2.3. Brokerage..................................................................................... 21 2.2.2.4. Structural Holes ........................................................................... 22 2.3 Summary........................................................................................................ 23 3. Research Methods ........................................................................................... 25 3.1. Research Direction of Study 1...................................................................... 25 3.1.1. Hypotheses.......................................................................................... 25 3.1.1.1. Mediation Effects of Knowledge Contribution............................. 25 3.1.1.2. Features of Personal Social Networks......................................... 26 3.1.1.3. Effects of Structural Holes ........................................................... 28 3.1.2. Instruments.......................................................................................... 29 3.2. Research Direction of Study 2...................................................................... 30 3.2.1. Hypotheses.......................................................................................... 30 3.2.2. Instruments.......................................................................................... 31 v.

(10) 3.3. Participants ................................................................................................... 32 3.3.1. Participants of Study 1 ........................................................................ 32 3.3.2. Participants of Study 2 ........................................................................ 32 3.4 Variables........................................................................................................ 32 3.4.1. Variables of Study 1 ............................................................................ 32 3.4.1.1. Relational Variables of Personal Relationships........................... 32 3.4.1.2. Research Variables....................................................................... 34 3.4.2 Variables of Study 2 ............................................................................. 35 3.4.2.1. Variables of Knowledge Acquiring .............................................. 35 3.4.2.2. Research Variables....................................................................... 36 3.4.3. Other variables .................................................................................... 37 3.5 Research Procedures...................................................................................... 37 3.5.1. Procedure of Study1............................................................................ 37 3.5.2. Procedure of Study 2........................................................................... 38 3.6 Data Collection.............................................................................................. 38 3.6.1. Data Collection of Study 1.................................................................. 38 3.6.2. Data Collection of Study 2.................................................................. 39 3.7. Data Analysis................................................................................................ 39 3.7.1. Data Analysis of Study 1..................................................................... 39 3.7.1.1. Relational Properties of Personal Networks ............................... 39 3.7.1.2. Two QAP Analyses of Personal Networks.................................... 40 3.7.1.3. Knowledge Prestige ..................................................................... 41 3.7.1.4. Effects of Structural Holes ........................................................... 42 3.7.2 Data Analysis of Study 2...................................................................... 42 3.7.2.1. Relational Properties of Personal Networks ............................... 42 3.7.2.2. QAP Analysis of Personal Networks............................................ 43 3.7.2.3. Knowledge Brokerage Scores of STCs in E-Schools ................... 43 4. Results .............................................................................................................. 45 4.1. Results of Study 1......................................................................................... 45 4.1.1. Demographic Descriptions.................................................................. 45 4.1.2. Relational Properties........................................................................... 45 4.1.3. QAP Correlations................................................................................ 46 4.1.4. Work Positions of Correlates .............................................................. 47 4.1.5. Knowledge Prestige ............................................................................ 47 4.1.5.1. Effects of PSN size........................................................................ 48 4.1.5.2. Effects of PSN density .................................................................. 49 4.1.5.3. Effects of personal tie strength..................................................... 49 4.1.5.4. Knowledge contributions and knowledge prestige ...................... 49 vi.

(11) 4.1.6. Social Configurations across School Boundaries ............................... 50 4.1.7. Effects of Structural Holes.................................................................. 53 4.2. Results of Study 2......................................................................................... 54 4.2.1. Demographic Descriptions.................................................................. 54 4.2.2. Relational Properties........................................................................... 55 4.2.3. QAP Correlations between Different Types of PSNs ......................... 55 4.2.4. Knowledge Brokerage ........................................................................ 55 4.2.5. Knowledge Brokerage of STCs in E-Schools..................................... 56 5. Discussions ....................................................................................................... 59 5.1. Knowledge Prestige...................................................................................... 59 5.1.1. Characteristics of knowledge contribution ......................................... 59 5.1.2. Characteristics and deficiencies of PSNs............................................ 60 5.2. Knowledge Sharing across School Boundaries............................................ 62 5.2.1. Configurations of Social Network ...................................................... 62 5.2.2. Deficiency of Knowledge Sharing of ICT-in-education..................... 63 5.3. Opportunities of Brokering Knowledge across School Boundaries............. 64 6. Conclusion & Recommendations for Future Studies .................................. 67 References............................................................................................................ 69 Appendix A. Questionnaire items of the studies .............................................. 81 Appendix B. Questionnaire of the First Study ................................................. 87 Appendix C. Questionnaire of the Second Study........................................... 103 Appendix D. Tables of Statistical Data ........................................................... 113. vii.

(12) List of Tables Table 3.1. Summaries of Relational Variables in Two Layers ..................................... 33 Table 3.2 Variables of Knowledge Brokerage and Friendship..................................... 36 Table 3.3. Composite Reliability (CR) and Average Variance Extracted (AVE) ........ 41 Table 4.1. QAP Correlations between Five Personal Knowledge Exchange ............... 46 Table 4.2. QAP Correlations between the Four Types of Knowledge Networks across School Boundaries ...................................................................................... 46 Table 4.3 Distribution of Correlates’ Work Positions in Five Knowledge Sharing ..... 47 Table 4.4. Goodness-of-fit Model................................................................................. 48 Table 4.5. The Direct, Indirect, and Total Effects of Knowledge Exchange on Prestige ..................................................................................................................... 49 Table 4.6. Results of the Multiple Regression Analysis for Predicting Prominence.... 54 Table 4.7. QAP Correlations between Five Types of Knowledge Acquiring Networks ..................................................................................................................... 55 Table 4.8. Correlations of Four Knowledge Brokerage Scores and STCs in E-Schools ..................................................................................................................... 57 Table 4.9. Correlations of Brokerage Scores and Friendship for STCs in E-Schools .. 57 Table A.1. Questionnaire Items (English Version) of the First Study and Corresponding Relational Variables ........................................................... 83 Table A.2. Questionnaire Items (English Version) of the Second Study and Corresponding Relational Variables ........................................................... 85 Table D.1. Statistical Descriptions and Correlations of Personal Relational Variables in the First Layer Interview....................................................................... 115 Table D.2. Statistical Description of Individual’s Relational Variables in the Second Layer ......................................................................................................... 116 Table D.3. Statistical Description of Relational Variables for Knowledge Acquiring and Friendship........................................................................................... 117 Table D.4. Statistical Brokerage Scores of Five Types of Knowledge ...................... 118. i.

(13) Lists of Figures Figure 2.1 Five types of Relational Brokerage 22 Figure 2.2 Example of a Triad 23 Figure 2.3 Four types of Relational Brokerage for STCs of E-Schools 31 Figure 3.1 Two-layer Interview Structure 29 Figure 3.2 Analytical Model of Prestige and Knowledge Exchange 34 Figure 3.3 Effects of Structural Holes on Prestige 35 Figure 4.1 Result of Hypothesis Testing 48 Figure 4.2 Graphic Drawing of Technology Knowledge Acquisition Networks 51 Figure 4.3 Graphic Drawing of ICT-in-education Knowledge Acquisition Networks 51 Figure 4.4 Graphic Drawing of ICT Knowledge Contibution Networks 52 Figure 4.5 Graphic Drawing of ICT-in-education Knowledge Contribution Networks 52 Figure 4.6 Result of Hypothesis Testing of ICT Practices 54 Figure 4.7 Result of Hypothesis Testing of ICT-in-education Practices 54 Figure 4.8 Average Scores of Knowledge Brokers for Different Knowledge 56. ii.

(14) 1.

(15) 1. Introduction 1.1 Statement of Problem Scholars. often. distinguish. practical/procedural. knowledge. from. theoretical/declarative knowledge. Polanyi (1966) argued that theoretical knowledge can be derived from reflection and abstraction, from experience. He indicated that all knowledge is either tacit or rooted in tacit knowledge. Tacit knowledge is not easily codified for reuse or sharing. Reber (1993) has argued that the acquisition of tacit knowledge “takes place largely in the absence of explicit knowledge about what was acquired.” It depends largely on frequent interaction and observation, similar to what takes place between an apprentice and a master. Polanyi (1966) claimed that tacit knowledge is not private but social. He placed strong emphasis on dialogue within an open community as a means of blending socially conveyed knowledge with the individual’s experience of reality. However, Bourdieu (1977) argued that an objective analysis of practical apprehension of the familiar world is beset by limitations. He suggested that researchers escape from the dichotomy of knowledge and prepare to “inquire into the mode of production and functioning of practical mastery.” He proposed a sort of knowledge that “conserves and transcends objectivist knowledge by integrating the truth of practical experience and of the practical mode of knowledge which this learned knowledge has to be constructed against the truth of all learned knowledge.” He argued that researchers shall “subordinate all operations of scientific practice to a theory of practice and of practical knowledge.” Bourdieu’s theory of practice laid the foundation for studies on work practice and experience in organizations. Successive scholars argued that knowledge sharing in relation to work was a process of situated learning and could be viewed as a social practice associated with designated 1.

(16) legitimate peripheral participation (Brown & Duguid, 1991; Lave & Wenger, 1991). They proposed a concept of community of practice (CoP) to refer to the privilege site of workers wherein there is a tight and effective loop of insight, problem identification, learning, and knowledge production (Brown and Duguid 1991; Lave & Wenger 1991). The CoP is recognized as an efficient and effective social configuration for facilitating knowledge sharing among professions (Brown & Duguid 2001; Wenger 2002). It allows workers to pursue competencies and communal practices (Barley 1996; Wenger 1998). It also mediates individuals and social structures, thus allowing communication between unacquainted practitioners. As network technologies have been deployed, concepts of CoPs have been extended (Teigland 2003). The advantage of an online community of practice (O-CoP) lies in its facilitating inter-organizational cooperation. The O-CoP enables distant workers to share expertise and to engage, collaborate, and build relationships through networks. In contrast to a face-to-face CoP in which the members’ participation is usually regulated by an authority, an O-CoP is sustained by the participants’ spontaneous contribution. Thus, much effort has been spent on finding methods to enhance online interactions (Spitzer & Wedding, 1995; Kling & Courtright, 2003; Sherer, Shea, & Kristensen, 2003), on creating strategies for facilitating reciprocal relationships (Schlager, Fusco, & Schank, 1998; Barab, MaKinster, & Scheckler, 2003; Saint-Onge & Wallace, 2003), and on designing frameworks for guiding O-CoP activities (Corbin, 2003). To permit a better understanding of the properties of online social behaviors, an O-CoP had been built and moderated in the first stage of research, called ASTC (Association of School Technology Coordinators), with a view to creating an open discussion space for ICT teachers over a period of two years. The ASTC focuses on promoting school technology integration (Lin, Pan, & et. al., 2004). After having 2.

(17) implemented several experimental facilitation strategies, the ASTC had more than 300 participants and had become a geographically dispersed and sparsely knit O-CoP (Lin, Hsu, & Chiou, 2005). O-CoP participation involves participants’ shifting from outsider membership to that of insider, and participants who visit repeatedly are more likely to become insiders (Wenger, 2002; Takahashi, Fujimoto, & Yamasaki, 2003; Rafaeli, Ravid & Soroka, 2004). In the study of ASTC, the authors found that the repeat-visiting frequency distribution of online participants can be characterized by a stochastic NBD model (NBD: negative binomial distribution). Though facilitations could gather participants, ICT teachers took part in discussions on their own initiative rather than as results of relational binding (Lin, Hsu, & Chiou, 2009). The results of ASTC studies revealed that a sparsely knit O-CoP had low repeat-visiting frequency; the studies also indicated the obstacles that must be overcome for the STCs’ community to be sustainable over the long term. 1.1.1 Social Networks underlying Social Practices Social practice emphasizes “the relational interdependency of agent and world, activity, meaning, cognition, learning, and knowing” (Reckwitz 2002). When encountering problems, workers usually rely on social networks for knowledge acquisition (Allen, 1977; Cross et al., 2001b). Social networks are demonstrably important for obtaining information, solving problems, and learning how to do work better (Cross et al., 2001a; Granovetter 1973; Levin and Cross 2004; Wenger 1998). Computer networks that link people, organizations, and knowledge are also inherently social networks (Wellman 2001). Sharing knowledge with other workers of similar profession is akin to participating in an open CoP, which is sustained by both face-to-face contacts and computer network communications. The author argues that while site functionalities, site facilitations, and members’ attitudes and expectations 3.

(18) were all influential, it is mainly participants’ social network with regard to knowledge sharing that determine extensive participation in an online CoP. A successful online CoP should become part of members’ personal social network (PSN) of knowledge sharing. Hence, knowledge sharing strategy is limited by an individual’s social network that encompasses both physical and virtual relationships. Knowledge seeking encompasses a social process of evaluating the awareness, cost, and value of helpers (Borgatti & Cross, 2003). In the open CoP, workers’ linking to prestigious colleagues may reduce their knowledge-seeking efforts. Our experience of managing ASTC also proved the role of socially prestigious participants. Prestige—indicating the degrees of connections and relationships through which others can gain access to information in a sufficiently timely fashion (Knoke & Burt, 1983)—might represent the advantage of knowledge sharing. However, while previous research on social practice merely emphasizes the central groups in participation processes, not many empirical studies focus on their underlying social networks. Today, workers extensively develop PSNs across organizational boundaries by use of the Internet. Support for knowledge acquisition across organizational boundaries is readily available; however, the Internet can provide codified knowledge, reusable work products, and communicated media while failing to enable workers to truly share knowledge (McDermott, 1999; Cross et al., 2001b). Lack of reciprocity will lead to the termination of a relationship (van Tilburg, van Sonderen & Ormel 1991). Weak binding between online participants makes knowledge sharing unbalanced and produces a loosely coupled system (Brown & Duguid, 2001) in which knowledge sharing is similar to a gift exchange, with an emphasis on the timing of a transaction, as discussed in Bourdieu’s theory of practices (1977). Reciprocation of 4.

(19) knowledge, like giving a gift in return, may be delayed and different in kind, or absent altogether. Nonetheless, some still actively contribute to knowledge, regardless of reciprocation from others—in particular, contributing knowledge to distant or unfamiliar participants. People who often contribute knowledge to community participants will take on central roles (Wenger, 2002). Central persons would attain higher social status and acquire prestige (Knoke & Burt, 1983). Researchers who seek intrinsic benefits in the virtual world have also emphasized gift-giving attitudes toward knowledge contribution, indicating that online social status is determined by what you give away, not what you control (Kollock, 1999; Raymond, 2003). That is, those participants engaging more in knowledge contributions will gain higher prestige in a loosely coupled system. Prestigious workers are assumed to have beneficial social networks. Because PSNs will affect the access of other individuals and the effectiveness of informal personalization, investigating the characteristics of PSNs of prestigious participants might serve to disclose the mode of these participants’ practical knowledge. 1.1.2 Social Resources of STCs’ Community Social network models conceptualize social structure in terms of relational processes, making relations that define linkages among people or organizations a fundamental component (Wasserman & Faust, 1994). Researchers have indicated that PSNs have features that may be used to promote self-interests (Lin, 2001), affect the outcome of job searches, and gather information (Granovetter, 1973; Marsden and Hurlbert, 1988). Social resources have been used to obtain network benefits that are accessible through social connections (Burt 1992; Granovettor, 1977). In an educational context, researchers emphasize the practical knowledge that is shared through informal relationships among teachers (Frank, Zhao, & Borman, 2004; 5.

(20) Carmichael, et. al., 2006). Scholars have argued that colleague-to-colleague networks in school are important for teachers’ professional development (Bidwell 2001; Tabert & McLaughlin, 1994). The social processes of knowledge transferal and the social relationships of knowledge sharing have positive impacts on educational innovations (Frank, Zhao, & Borman, 2004; Patrick, et. al., 2004; Muthukumar & Hedberg, 2005). School technology coordinators (STCs) are ICT teachers who are responsible for all ICT maintenance and are leaders of educational technology innovations. Case studies of STCs have indicated that STCs are keen to acquire exemplary practices with regard to ICT use and ICT-in-education (Evans-Andris, 1995; Lai & Pratt 2004; Lesisko 2005; Lin, Pan & Chiou 2004) and consequently seek out economical ways of exchanging knowledge (Dexter, Anderson, & Ronnkvist, 2002). Through network technology and building a knowledge repertoire (McAndrew, et al. 2004; Carmichael and et. al. 2006), STCs can easily gain support across school boundaries. However, researches have also addressed the barriers imposed by STCs’ social-structural conditions (Evans-Andris, 1995; Marcovitz, 2000; Lai, Trewern, and Pratt, 2002; Lai, et. al 2004; Place and Lesisko, 2005) and the lack of professional development (Lai & Pratt, 2004). Bryderup and Kowalski (2002) have argued that support for ICT-in-education requires that the municipality perspective be followed up on. They claimed that “the schools’ responses to municipal ICT policy aims, and actions to meet these aims, concern the distribution of resources (Bryderup & Kowalski, 2002).” In Taiwan, much effort has been expended on building online communities for ICT teachers. Thanks to their ICT competencies, STCs performed important roles on these projects. That is, STCs of ICT-in-education exemplar schools are typically the primary coordinators to share practical knowledge and experience with other STCs. They probably have better social resources for knowledge transmission than do STCs of 6.

(21) other schools. While Carmichael et al. (2006) used a mapping technique to explore the personal network benefits available to other colleagues, there are still few quantitative studies that explore the network benefits embedded in teachers’ knowledge sharing.. 1.2 Research Purposes This research had three objectives. The first was to explore the exchange characteristic that marks knowledge sharing. Prestige’s usefulness as an indicator of knowledge sharing in an open CoP was verified. Whether knowledge contribution can mediate between limited informal personalization and prestige in a loosely coupled CoP was also identified. The second was to investigate the effective social resources of informal personalization of knowledge sharing. We investigated the sets of relationships across school boundaries, which STCs rely upon for solving ICT and ICT-in-education problems. We also inspected the differences seen between informal personalization when sharing ICT-in-education practices and that when sharing ICT practices. The third objective was to explore the opportunities for information brokering that are available due to STCs’ knowledge acquisition interests. We intended to classify the characteristics of brokering, and to explore the opportunities for brokering various practices of ICT and ICT-in-education among STCs of government-funded exemplar schools (E-School) and STCs of other schools.. 1.3 Operational Definitions. CoP Correlates Effective network sizes Ego-centric network. Abbreviation of Community of Practice Correspondents in specific relationships with STCs Number of non-redundant links to correlates “Partial network” that is anchored around a particular individual, i.e., personal social network 7.

(22) Friendships. STCs of other schools perceived by the subject as good personal friends. ICT-practice Work related to ICT in school contexts. ICT-in-Education-practice Work related to integrating ICT with educational practices. Knowledge acquiring Knowledge acquisition or attainment in solution-seeking contacts. Knowledge brokerage Positional advantages of socially brokerage in knowledge sharing. Knowledge contribution Knowledge giving or providing in solution-seeking contacts. Knowledge exchange Knowledge sharing involving knowledge acquisition and reciprocation, Knowledge prestige Positional advantages of socially prestigious persons in knowledge sharing. Knowledge sharing Indicating the activities of knowledge exchange or knowledge contribution in solution seeking contacts. Knowledge reciprocating Reciprocating knowledge to helpers. O-CoPs Abbreviation of Online Community of Practices Personal knowledge Ego-centric networks of knowledge sharing sharing PSN Abbreviation of personal social network Practice Habitual performance or work of professionals. Relational properties Social network characteristics of personal relationships. Structural hole A network location which connects to many correlates that are themselves unconnected Tie A relationship (or a link) to a correlate.. 1.4 Structure of the Dissertation This dissertation is structured as follows. Chapter One describes the research problems, research purposes, research questions, and definitions of terms. Chapter Two reviews literatures pertaining to the theoretical and methodological foundations of social practice and social networks. Chapter Three describes research directions, participants, instruments, data collection, and analysis. Chapter Four shows the results. Chapter Five discusses the findings and the implications. Chapter Six concludes this 8.

(23) research and discusses directions for future research. The Appendices A to C list the questionnaires and interview items used in this research. Appendix D lists tables of statistical description and correlations of relational variables.. 9.

(24) 10.

(25) 2. Literature Review 2.1. Theoretical Foundations Social practice and social network theories provide the theoretical foundations to support our arguments of knowledge sharing. Bourdieu’s theory of practices (1977) envisaged apprehension as taking place in a practical universe that was situated in a practical space. He argued that social behavior is “a continual accomplishing of actions in the implementation of natives’ strategies in accordance with their practical mastery of situations.” He also proposed the notion of habitus to capture “the permanent internalization of the social order” (Bourdieu, 1990) and proposed the agent’s practice as “his or her capacity for invention and improvisation.” Postill (2001) has indicated that Bourdieu’s notion of “fields” indicates “specialist domains of practice with their own ‘logic’ that are constituted by a unique combination of species of capital, e.g., financial capital, symbolic capital (prestige, renown) or social capital (connections).” Concepts of “fields” and “habitus” provided the foundations of researches about knowledge sharing. Brown and Duguid (1991) have indicated that knowledge is enhanced from the perspective of practice. Practice is “a way of talking about shared historical and social resources, frameworks, and perspectives that can sustain mutual engagement in action (Wenger, 1998, pp. 5).” Practice also circulates knowledge and transfers tacit knowledge explicitly (Brown & Duguid 2001). Lave and Wenger (1991) argued that knowledge sharing of practices was a process of situated learning and could be viewed as social practices associated with designated legitimate peripheral participation. Reckwitz (2002) has indicated the importance of social practices as bodily and mental routines of practice theories. Social practice emphasizes “the inherently socially negotiated character of 11.

(26) meaning and the interested, concerned characters of the thought and action of persons-in-activity (Lave & Wenger 1991).” The CoP was proposed to refer to a kind of privileged site with a tight and effective loop of insight, problem identification, learning, and knowledge production in organizations (Brown & Duguid, 1991). Lave and Wenger (1991) indicated that “learning as increasing participation in CoPs concerns the whole person acting in the world.” In this decade, Brown and Duguid (2000) have proposed a term “network of practices (NoPs)” to indicate the social space of an electronic network where individuals working on similar problems have self-organized to help each other and share perspectives about their occupational practice or common interests. Teigland (2003) argued that NoPs range from communities of practice to electronic networks of practice. Participation in NoPs means engaging in knowledge acquisition and contributing to work practices, whether via face-to-face or Internet techniques. Consequently, the strategy of knowledge sharing is limited by an individual’s social network that encompasses physical and virtual relationships. 2.1.1. Network Benefits of Knowledge Sharing The social network of an individual consists of his/her informal interpersonal relations. Researchers indicated that social factors in networks, such as co-location (Allen, 1977; Kraut et al. 1990, Cummings, 2004), demographic similarity (Pelled 1996), and group member diversity (Cummings, 2004), as well as the relational properties (Wellman & Wortley 1990; Cross and Cummings, 2004) have important effects on knowledge sharing (Cross, Rice & Parker, 2001b). The use of social contacts with colleagues and the resources provided by the contacts positively affect individuals’ work performance (Cross & Cummings, 2004) and knowledge creation (Teigland & Wasko, 2000). An individual’s advantage of knowledge exchange is based on his or her control 12.

(27) over the spread of information. Cross and Cummings (2004) have verified that the number of acquisition ties significantly affects knowledge workers’ performance ratings. Friedkin’s research (Friedkin et al. 1994) indicated that a cohesive network among teachers enhances their performance. While people with high cohesive networks are linked by other factors besides interaction and perceived similarity, higher network density would decrease the effects of the knowledge transmittal between people (Burt 1992). McFadyen and Cannella’s (2004) study has indicated that strength of interpersonal relations had a higher marginal effect on knowledge creation. While Lin (2001) has proposed that weaker ties facilitate status attainment, Teigland and Wasko (2000) indicated that the perception of a worker’s having stronger relationships with collocated colleagues is negatively associated with knowledge innovation (Teigland & Wasko 2000). 2.1.2. Gift-Giving Attitudes of Knowledge Sharing Network ties, created through knowledge sharing, include relationships of knowledge acquisition and knowledge contribution. Reciprocity is critical to maintaining the intimate relationships of exchanges (Nahapiet & Ghosha 1998). Participants receive meaning retrospectively from the responses they generate at some point in the future. Lack of reciprocity will lead to the termination of a relationship (van Tilburg, van Sonderen & Ormel 1991). However, some participants voluntarily provide a benefit to others with the expectation of some future return, even though the exact nature of the return is unclear. Bourdieu (1977) argued that a critical point of exchange is one that “constitutes as reversible a practice that agents construe in performance as irreversible.” He proposed a concept of gift exchange and emphasized its playing on the timing or tempo of a transaction. He indicated that “it is all a question of style, which means in this case timing and the choice of occasion; for the same act—giving, giving in return, 13.

(28) offering one’s service, paying a visit, etc.—can have completely different meanings at different times, coming as it may at the right or wrong moment (pp. 6).” Researchers who seek intrinsic benefits in the virtual world have emphasized gift-giving attitudes toward knowledge contribution (Rheingold 1993; Kollock 1999). People engage in social contribution with the expectation of social rewards (Burt, 1992; Nahapiet & Ghoshal, 1998). Though reciprocation is not an explicit guarantee of equivalent-value feedback at a future date, people can through their contributions acquire personal resources that represent symbolic capital, which is used to gain power or social status (Lin, 2001). Individuals contribute knowledge to receive intrinsic benefits from a self-evaluation of the activity itself, rather than to receive external rewards (Wasko & Faraj, 2005). In other words, people contribute knowledge because they feel good in doing so (Kollock, 1999; Wasko & Faraj, 2005). Gaining support from network members (Plickert et al., 2007), recognition among peers, and highly valued aspects of professional identity (Lamb and Davidson, 2005) are significant motivations for knowledge contribution. Kollock (1999) has argued that the favors and benefits provided in online communities are public goods, i.e., goods from which all may benefit. A command hierarchy and exchange economy is the basis for the conventional principles governing the cultivation of a community. Recently, Rheingold (1993) has described the interactions within one online community as a gift economy and Raymond (2003) has applied gift cultures to describing online behaviors. A gift culture assumes that online social status is determined not by what you control but by what you give away (Raymond, 2003). Gift cultures are adapted not to scarcity but to abundance. This abundance makes reputation among one’s peers the only available measure of competitive success (Raymond 2003).. 14.

(29) 2.1.3. Positional Advantage of Knowledge Sharing The primary stage of knowledge acquisition involves knowing what another person knows and being able to gain access to that person (McDermott, 1999; Cross et al., 2001a; Borgatti & Cross, 2003). Since people who encounter problems contact experts who have solutions, knowing who has expertise is an important factor that dominates support-seeking behavior (Lin & Chiou, 2008). Who is prestigious when engaging in knowledge exchanges? A participant about whom other participants express positive sentiments when recalling knowledge acquisition is a prestigious person in their community (Lin 2001). That is, prestigious people are major channels of relational information who maintain a large number of direct contacts with or adjacent to many other actors (Burt, 1992; Wasserman & Faust, 1994). Greenberg (1964) indicated that in the pre-Internet era, individuals who occupied a central location in a social network could easily reach others for personal discussion and could be easily reached themselves. In recent decades, researchers have shown that a person moves closer to the center as he or she becomes more involved in team activities through computer networks (Bradner et al., 1998). Lai and Wong (2002) also argue that central people are found to be active and efficient in transmitting information within the group. Central people (or prestigious people) epitomize the advantage of knowledge sharing. As knowledge exchanges through virtual work become more common, researchers have indicated that online social status is determined by what you give away, not what you control (Kollock, 1999; Raymond, 2003). Prestige has become a measure of online success. Walsh and Bayma (1996) also indicate that computer-mediated communication (CMC) provides new opportunities and resources to scientists located at less-prominent institutions; they argue that those scientists can gain prestige through the Internet. People who often contribute knowledge to 15.

(30) community participants will take on central roles (Wenger, 2002). Central persons would attain higher social status and prestige (Knoke & Burt, 1983). The influence of boundary spanning on organizational learning has also been investigated (Burt, 1992). Ties crossing organizational boundaries are positively related to organizational competitiveness (McEvily & Zaheer, 1999) or individual performance (Cross and Cummings 2004). Wenger (1998) indicated that people in charge of special projects across functional units often find themselves brokering. Fielding et al. (2005) argued that sharing work across the school with everyone in it can maximize the chance of good practice being transferred internally (Fielding et al. 2005). Socially brokerage persons of knowledge sharing networks have positional advantages (or social resource benefits) in contacts and have opportunities of knowledge brokerage.. 2.2 Methodological Foundations Researchers have indicated that “the network perspective allows new leverage for answering standard social and behavioral science research questions by giving precise formal definition to aspects of [the] social structural environment (Wasserman & Faust 1994).” Social network models conceptualize social structure as patterns of relations (Wasserman & Faust, 1994). Relations that define linkages among people or organizations are fundamental components of social network theories (Wasserman & Faust 1994). Relational ties are channels for transfer or flow of resources (Burt 1992; Granovettor, 1973). Social network researchers have suggested that “developments in social networks analysis can point the way to novel frameworks of sociological theory” (Scott, 2001.). Social network researchers explored network processes through the use of frequency tabulations and made qualitative comments on the structure of the network 16.

(31) relations that they discovered (Scott, 2001). In his research on job-seeking, Granovetter has proposed an information diffusion model and has argued that the acquisition of information depends on those with information having the motivation to pass it on and the strategic location of a person’s contacts in the overall flow of information (1973). These concepts of network benefits have been synthesized as social resources that are accessible through social connections. 2.2.1 Social Network Analysis Social network analysis (SNA) assumes that relationships among interacting units are important (Wasserman & Faust, 1994). The unit of SNA is an entity consisting of a collection of individuals and linkages among them. Mitchell (1969) has argued that it is necessary to identify particular aspects of the total network in conceptualizing the total network of a society. He has devised the concept of a “partial network” that is anchored around a particular individual so as to generate an “ego-centric” network of social relations. He also devised the concept of a “partial network” by the content or meaning of the relations involved. From thenceforth, partial network studies were always ego-centered networks and focused on particular types of social relationships (Scott, 2001). These networks are multiplex and involve the combination of a number of meaningfully distinct relations (Scott, 2001). The following sections describe the foundations of content analysis and ego-centric network analysis. 2.2.1.1 Network Content In SNA, relations are analytical constructs. Network analysts commonly focus on interaction between network effects and contents. Podolny and Baron (1997) proposed that the content of networks in the workplace affects individual mobility. Burt’s theory on social ties of information and resources reflects a highly instrumental view of networks (Burt 1992). 17.

(32) Burt (1983, pp. 36) indicated that “network models of social structure typically describe the form of relations while taking the content of those relations as a given, an item exogenous to the model.” Relations’ forms are described by their intensity or strength of occurrence and relations’ contents are described by the substance of occurring. Based on the analytical objective, the model requires that subjects “interpret specific interaction activities as manifestations of more general types of activities (1983, pp. 66).” Granovetter also argued that “social embeddedness of economic exchange overlaps with networks that convey resources-based and identity-based content” (1985). 2.2.1.2. Ego-centric Network Analysis An individual’s social network of informal interpersonal relations (or ties) is called a PSN or an ego-network. PSNs have been conceptualized as representations of the informal personalization and potential influence of individuals (Wasserman and Faust, 1994; Degenne and Forse, 1999; Stefanone et al., 2004). Burt (1992) indicated that “it is more useful analytically to focus on the pattern of relationships among the people to whom ego is tied.” The size, strength, and density of personal networks are usually used to illustrate the essential phenomenon of an ego-centric network (Cross & Cummings, 2004; Teigland & Wasko, 2000; Wasserman & Faust, 1994; Wellman, 2001). Ego-centric network surveys are often conducted through subjects’ recall and name generation (Wasserman & Faust, 1994). Subjects perception of their ego-centric networks reflects their relational resources. Researchers have argued that people are generally very inaccurate in reporting on their past interactions but they also showed that people remember long-term or stronger relationships (Freeman et al., 1987; Corman & Bradford, 1993). Richer data can be obtained with the free recall method while the fixed choice method can provide accurate information on the most important relationships (Hammer, 1984; Hlebec, & 18.

(33) Ferligoj, 2002). Kogovsek and Ferligoj (2005) have verified that behavioral questions were more reliable than questions with emotional content. Ego-centric networks have been widely used to study the social structures surrounding individuals and are sometimes conceptualized as representations of the potential influence of individuals (Stefanone et al. 2004; Wasserman & Fraust 1994). Researchers have extended the ego-centric approach to investigating the transactions or exchanges of information (Granovetter, 1973, Burt, 1992), services, and affection (Degenne and Forse 1999). Cross et al. have proven that interdependence of tasks is the important predictor of information-seeking through personal relationships in organizations (Cross, Rice, and Parker, 2001b). 2.2.2. Social Resource Social resources are resources accessed through an individual’s social connections (Campbell and Marsden 1986). Researchers of social resources argued that the features of a personal network include opportunities to promote that one’s self-interests (Lin 2001), affect the outcome of job searches (Granovetter 1995; Marsden and Hurlbert 1988), and gather information. Resources have different values in human groups or communities (Sewell 1992; Lin 2001). Lin (2001) proposed “the community promotes its self-interest by conferring relatively higher statuses on individuals who possess valued resources.” In social networks, researchers argue that fluidity characterizes the occupants, positions, resources, and procedures (Lin 2001) Social resource research typically asks how the characteristics of the contacts that provided information or the nature of the contact tie affected outcomes (Marsden and Hurlbert 1988; Lin, Ensel and Vaughn 1981; Lin 2001). For example, Burt (1992) proposed a concept of structural holes (i.e., being connected to many actors who are themselves unconnected) and verified that structural holes can enhance career opportunities for workers who are competing for 19.

(34) promotions. Network locations will create competitive advantages by linking activated ties to outcomes (Burt 1992). 2.2.2.1. Relational Properties of PSNs Measurements on the ties among actors are relational and differ from standard social science data. The level at which network data are studied is referred to as a modeling unit. Network ties, created through social relationships, consist of behaviors of receiving and reciprocating. Network analysts often collect network data by observing, interviewing, or questioning individual actors about the ties they have with other actors. The dyad is the common unit of observation (Wasserman & Fraust 1994, pp. 43). Number of contact ties, network density, and tie strength affect the effectiveness of informal personalization. The pattern of ties and the relationships built on them are the foundations of social capital (Nahapiet & Ghoshal 1999). In turn, social capital facilitates exchange (Nahapiet & Ghoshal 1999). Socioeconomic statuses are positively related to network size (Compbell, Marsden, & Hurlbert 1986) and contact resources (Lin 1988). The number of acquisition ties is significantly related to knowledge workers’ performance ratings (Cross & Cummings, 2004). Frank, Zhao, and Borman (2004) have further proven that informal accesses to expertise are manifestations of social capital that can facilitate the implementation of innovation. Network density is another common measurement of informal personalization. It is calculated by counting the number of ties that connect alters and then dividing the sum by the number of pairs (Wasserman & Faust 1994). Burt (1992) has indicated that people with a low density of ties have fewer constraints and perform better. On the other hand, a dense relationship can result in secure relationships, which would further facilitate cooperative intention (Nahapiet & Ghoshal 1998; Jameson, et al., 2006), high trust (Buskens, 1998) and assist in solving problems (Sparrowe et. al. 20.

(35) 2001). Tie strength has been measured in various ways and frequency of contact was one popular indicator. Strong ties indicate frequent contacts that almost invariably are affective and reciprocal (Granovetter 1973). While strong ties are for solidarity, weaker contacts, that is, infrequent contacts, are sources of new information and are important for the dispersion of information (Burt 1992; Granovetter 1973). Strong ties are bridges of information pertaining to the interest of a group and/or its individual members (Lai and Wong 2004) and are negatively associated with knowledge innovation (Teigland & Wasko, 2000). Weak ties tend to be bridges to different social circles and are embedded resources that are both heterogeneous and useful (Burt, 1992; Granovetter, 1973; Lin, 1982; Montgomery, 1992), which facilitate personal status attainment (Lin, 2001). 2.2.2.2. Centrality Degree centrality is commonly used to indicate positive recipients and positional advantages (Knoke & Burt, 1983; Wasserman & Fraust, 1994). An individual with a significant degree of direct contacts or an individual who is adjacent to many other correlates is considered a major channel of relational information (Wasserman & Fraust, 1994) and has prestige. Prestigious persons attract followers seeking long-term relations (Knoke & Burt, 1983, Wasserman & Fraust, 1994). Prestigious indices measure the directional relations because prestige is regarded as social status (Lin, 2001) and becomes salient, especially if the positive choices are not reciprocated (Nooy, Mrvar, & Batagelj, 2005). 2.2.2.3. Brokerage Researchers of social networks have defined a broker as a manipulator of people and information who provides the benefits of linking persons who would not otherwise be in communication (Burt, 1992). Brokers have opportunities to bridge 21.

(36) critical resources and have network effects on individual’s bargaining power and influence in dyadic exchanges (Burt, 1992; Fernandez & Gould, 1994). Gould and Fernandez had proposed that exchanges between some actors may have a different meaning or function from exchanges between other actors. They proposed that “brokerage should distinguish communication of resource flows within groups from flows between groups.” They proposed five structurally distinct types of brokers that follow from a partitioning of actors into non-overlapping subgroups (Gould & Fernandez 1989).. Figure 2.1 Five types of Relational Brokerages Although Gould and Fernandez’ theory (1989, 1994) of brokerage focused on the study of inter-organizational relations, the model provided a general, rigorous formulation of brokerage behavior: A “coordinator” brokers the resources with correlates within the same group (Fernandez & Gould, 1994). A second type of broker acts as a “gatekeeper” for a group and decides whether or not to grant access to an outsider. The third type of broker acts as a “representative” for a fellow group and attempts to establish contact with an outsider. A “consultant” is in a different group and acts as an itinerant broker for two correlates of the same subgroup. Finally, a “liaison” is an outsider who links distinct groups without having prior allegiance to either (see Figure 2.1). 2.2.2.4. Structural Holes Burt (1992) coined the term “structural holes” to refer to the positional advantage/disadvantage of individuals resulting from how they are embedded in 22.

(37) neighborhoods. In the (a) network of Figure 2.2, ego has an advantageous position as a direct result of the structural hole between actors A and B. That is, the ego is a strategic player and responsible for information benefits. He builds an efficient effective network and maintains the bridge tie (Burt 1992). The effective size of the network is given by the number of alters that the ego has minus the average number of ties that each alter has with other alters. Hence, the effective size of the ego of network (a) is greater than that of the ego of network (b).. Figure 2.2 Example of a Triad However, the network (b) of Figure 2.2 is more cohesive though the ego has a weaker integration advantage. That is, the connection between ego and ‘B’ is redundant because he can reach ‘B’ through ‘A’. A more cohesive network has more constraints for the person in the middle position (Nooy, Mrvar, & Batagelj, 2005, Wasserman & Faust, 1994).. 2.3 Summary A conceptual framework of this research is illustrated in Figure 2.3. This research explores the phenomena of knowledge sharing from a social practice theoretical perspective and investigates the social interdependency among knowledge workers. It proposes that knowledge prestige and knowledge brokering are positional advantages that come from knowledge sharing and can be examined by the measurement of social resources. It assumes that knowledge contribution facilitates the advantages that come from knowledge sharing and develops 23.

(38) hypotheses to verify the assumptions. Ego-centric network analysis is adopted for the purpose of conducting advanced analysis of the relational data.. Figure 2.3 The conceptual framework of this research. 24.

(39) 3. Research Methods In keeping with the research objectives and the conceptual framework, this research has developed two studies. The first study was designed to facilitate the first two research objectives. The second study was designed to facilitate the third research objective.. 3.1. Research Direction of Study 1 This study proposed that subjects who were identified more often by other subjects in a knowledge-acquisition network were more prestigious. Degree prestige (out-degree) measurement (Wasserman and Faust, 1994) by objective STCs (i.e., knowledge contributors) was used as the prestige score. 3.1.1. Hypotheses Theories of social practices and researches of network benefits, gift-giving attitudes, and positional advantages for knowledge sharing sustain the hypotheses of this study. Based on the arguments of network analyzers, this study uses ICT practices and ICT-in-education as the essence of knowledge sharing and proposed that prestigious persons take on central roles in knowledge dissemination. It explores the usefulness of prestige as an indicator in CoP and argues that modes of knowledge sharing can be clarified by analyzing the relational properties. The author develops assumptions to explain the positive effects of social resources on knowledge sharing. 3.1.1.1. Mediation Effects of Knowledge Contribution This study first seeks to verify the usability of prestige as an indicator of knowledge exchanges in a sparsely connected CoP. Then it seeks to identify whether gift-giving attitudes can mediate between limited informal personalization and prestige. It hypothesizes that STCs who engage in greater knowledge contribution to STCs at 25.

(40) other schools will have higher prestige. It also hypothesizes that STCs who are devoted to contributing knowledge to online communities could gain prestige. The range of a network refers to a degree of range that contains socially engaged network members. Hypothesis 1 (H1): Knowledge contribution to STCs at other schools has mediation effects for STCs leading to their gaining higher prestige scores. Hypothesis 2 (H2): Knowledge contribution to online communities has mediation effects for STCs leading to their gaining higher prestige scores. 3.1.1.2. Features of Personal Social Networks Features of personal social networks (PSNs)—e.g., number of contact ties, network. density,. and. tie. strength. within. organizations. or. among. inter-organizations—affect the effectiveness of informal personalization as a knowledge-exchange mechanism. In a school context, a knowledge-exchange PSN within or outside of school would reflect an STC’s informal personalization in seeking out support when encountering problems. The various features of PSNs’ underlying knowledge exchange would depict the varied status in CoPs. In the exchange process, network ties are created that consist of ties of knowledge acquisition and reciprocation. This. study assumed that STCs with large. knowledge-exchange PSNs both within and outside of school would thus have more resources for contributing knowledge to STCs at other schools. STCs with large knowledge-exchange PSNs would also tend to contribute more to online communities of related competency. Hypothesis 3 (H3): Size of STC’s personal knowledge-exchange network correlates with knowledge contribution to STCs at other schools. Hypothesis 4 (H4): Size of STC’s personal knowledge-exchange network correlates with knowledge contribution to online communities. Network density is calculated by counting the number of ties that connect members 26.

(41) of the network and dividing it by the total number of possible pairs in the network (Wasserman & Faust, 1994). This study argues that contributions to STCs of other schools depend more on a gift-giving attitude than trust. This study assumed that dense knowledge-exchange PSNs might not facilitate knowledge contribution to STCs at other schools. Hypothesis 5 (H5): Density of STC’s personal knowledge-exchange network does not correlate with knowledge contribution to STCs at other schools. This study assumed that an STC with a dense knowledge-exchange PSN engages in online knowledge exchange with community members because that STC prefers secure relationships. Hypothesis 6 (H6): Density of STC’s personal knowledge-exchange network positively correlates with knowledge contribution to online communities. Previous studies have used frequency of contact to represent tie strength. Because exchanges of tacit knowledge and complicated practices require frequent contact and greater effort to contribute to STCs at other schools, we assumed STCs with stronger knowledge-exchange ties might tend not to contribute knowledge to STCs of other schools. Hypothesis 7 (H7): Tie strength of STC’s personal knowledge-exchange network negatively correlates with knowledge contribution to STCs at other schools.. STCs with weaker knowledge-exchange ties presumably possess more heterogeneous information. Since contributing to an online community is easier than dealing with individuals and relatively heterogeneous enquiries from non-acquaintances, STCs that have PSNs with weak tie strength are assumed to be more likely to engage frequently in online relationships.. 27.

(42) Hypothesis 8 (H8): Tie strength of STC’s personal knowledge-exchange network negatively correlates with knowledge contribution to online communities. 3.1.1.3. Effects of Structural Holes The following hypotheses investigated the chance of good practice being transferred across school boundaries. This study inspected the efficiency of informal personalization of prestigious STCs by constructing a predictive model. The results provided the basis for discussing the discrepancies found between STCs’ PSNs in disseminating ICT practices and those in disseminating ICT-in-education practices Burt (1992) has proposed that a network with “structural holes” is efficiently effective. Greater effective sizes increase the efficiency of knowledge sharing and work performance. In contrast, high network density is hypothesized to have more constraints. That is, higher network density would decrease the effects of the knowledge transmittal to more people. This model hypothesizes that STCs who have larger effective network sizes of knowledge contribution will have higher prestige. It also hypothesizes that a lower density network with structural holes has fewer constraints and is positively linked to knowledge prestige. Hypothesis 9(H9): Effective network size of STC’s personal knowledge contribution network has positive effects on knowledge prestige. Hypothesis 10(H10): Density of STC’s personal knowledge contribution network has negative effects on knowledge prestige. Researches have shown that people with strong ties belong to cliques and strong ties tend to be located in or develop into cliques (Travers & Milgram, 1969). Contrastingly, weaker or irregular contacts are better sources of new information or are bridges to distant information networks. Weak ties are often more important for the dispersion of information than strong ties. Hypothesis 11(H11): The tie strength of STC’s personal knowledge contribution 28.

(43) network has negative effects on knowledge prestige. 3.1.2. Instruments This study developed a two-layer interview structure (Figure 3.1) with a two-layer interview protocol (see appendix A, Table A.1). The first interview regarding informal personalization was designed to examine the relational variables of each STC’s personal knowledge exchange. Correlates of informal personalization were colleagues with whom the subjects exchanged knowledge. The second interview regarding inter-school interaction was designed to record relationships of knowledge contribution to and acquisition from STCs of other schools. Researchers have indicated that workers’ recall of some specific interactions occurred at specific time intervals and has lower reliability than more general measures of typical interactions (Cross, Rice and Parker 2001). With this in mind, questionnaire items with general wording (not specific wording) were prepared for the two studies.. Figure 3.1 Two-layer Interview Structure. 29.

(44) 3.2. Research Direction of Study 2 This study proposed that subjects who were identified by other subjects in a knowledge-acquisition network were impressive knowledge contributors. The direction of study 2 is to investigate the brokering opportunities of knowledge contributors. This study also assumes the influence of friendship on STCs’ knowledge acquisition. The brokering scores will be compared with friendship relationships. To gain a thorough understanding of the whole network of practices, this study explored knowledge acquisition with regard to solutions for various ICT and ICT-in-education practices. These practices include computer hardware maintenance, software usages, network techniques, ICT procurement, and digitalizing materials. 3.2.1. Hypotheses The following hypotheses examine knowledge contribution relationships between different groups. That is, they explore the social coordinate roles of STCs for knowledge sharing. They inspect the characteristics of knowledge brokerage among STCs of government-funded ICT-in-education exemplary schools (E-School) and STCs of other schools, and explore discrepancies in their social resources (see Figure 2.3). They assume the brokered positions that can facilitate STCs of E-Schools in transferring innovative practices across school boundaries. The relational “brokerage notions” of Gould and Fernandez’s theory are used to study the opportunities for knowledge brokerage. This study examines four brokerage scores: coordinator, gatekeeper, representative, and consultant in STCs’ knowledge acquisition networks and focuses on ties among STCs of E-Schools and STCs of other schools. This study also examines several relative practices of ICT-in-education for the purposes of comprehending requirements for STCs’ professional development. The investigative practices include knowledge of computer hardware maintenance, software 30.

(45) usages, network techniques, ICT procurement, and digitalizing materials. This study compares the brokering opportunities presented by these practices with friendships among STCs. Hypotheses are described in the following. Hypothesis 12(H12): STCs of E-Schools have significantly higher coordinator scores of knowledge sharing than STCs at other schools. Hypothesis 13(H13): STCs of E-Schools have significantly higher gatekeeper scores of knowledge sharing than STCs at other schools. Hypothesis 14(H14): STCs. of. E-Schools. have. significantly. higher. representative scores of knowledge sharing than STCs at other schools. Hypothesis 15(H15): STCs of E-Schools have significantly higher consultant scores of knowledge sharing than STCs at other schools.. Figure 3.2 Four types of Relational Brokerage for STCs of E-Schools 3.2.2. Instruments In the second study, a survey (see appendix A, Table A-2) regarding inter-school interaction was designed to elicit information about relationships involving knowledge acquisition from STCs at other schools. Relational questions are used to ask STCs about their personal network of knowledge acquisition and friendship across school boundaries. According to the questions, participants select their choices from a list of names.. 31.

(46) 3.3. Participants 3.3.1. Participants of Study 1 In April 2007, the interviewers began to visit STCs of all junior high schools in Taipei City and conducted two in-depth ego-centric network interviews. A total of 39 out of 59 STCs participated in the interviews in the following two months. 3.3.2. Participants of Study 2 In January 2008, invitation letters were sent to the STCs of all junior high schools in Taipei County. STCs were asked to fill out the paper-based questionnaire, which was then mailed back. A total of 49 out of 59 STCs participated in the survey.. 3.4 Variables A large number of relational variables are defined to correspond to the ego-centric network interviews and surveys. They represent the raw network data and are computed for research variables of analytical models. 3.4.1. Variables of Study 1 3.4.1.1. Relational Variables of Personal Relationships CTKASize and CTKCSize represent variables of network size of ICT knowledge acquisition and the contribution of each STC in first layer. ATKASize and ATKCSize represent variables of network size of ICT knowledge acquisition and the contribution of each STC in second layer CIKASize and CIKCSize represent variables of network size of ICT-in-Education knowledge acquisition and the contribution of each STC in first layer. AIKASize and AIKCSize represent variables of network size of ICT-in-Education knowledge acquisition and the contribution of each STC in second layer. DISCSIze represents variable of network size of discussions for ICT-in-Education knowledge in first layer. CKADen and CKCDen represent variables of network densities of ICT knowledge 32.

(47) acquisition and contribution of each ego in work layer. AKADen and AKCDen represent variables of network densities of ICT knowledge acquisition and the contribution of each ego in second layer. CIKADen and CIKCDen represent variables of network densities of ICT-in-Education knowledge acquisition and the contribution of each ego in first layer. AIKADen and AIKCDen represent variables of network densities of ICT-in-Education knowledge acquisition and the contribution of each ego in second layer. CKAStr and CKCStr represent variables of tie strength of ICT knowledge acquisition and the contribution of each ego in first layer. AKAStr and AKCStr represent variables of tie strength of ICT knowledge acquisition and the contribution of each ego in second layer. CIKAStr and CIKCStr represent variables of tie strength of ICT-in-Education knowledge acquisition and the contribution of each ego in first layer. AIKAStr and AIKCStr represent variables of tie strength of ICT-in-Education knowledge acquisition and the contribution of each ego in second layer. IKCSize represents variable of number of networks spanning online knowledge contribution. The Table 3.1 summaries these relational variables. Table 3.1. Summaries of Relational Variables in Two Layers Direction of Knowledge Sharing. Knowledge Acquisition. Relational. Variable. Properties. (First Layer). (Second Layer). Network Size. CTKASize. ATKASize. CIKASize. AIKASize. CTKADen. ATKADen. CIKADen. AIKADen. CTKAStr. ATKAStr. CIKAStr. AIKAStr. Network Density Network Strength Effective Sizes 33. Name. Variable Name in. TKAEffSize.

(48) IKAEffSize Knowledge Contribution. Network Size Network Density Network Strength. CTKCSize. ATKGSize. CIKCSize. AIKGSize. CTKCDen. ATKGDen. CIKCDen. AIKGDen. CTKCStr. ATKGStr. CIKCStr. AIKGStr. Effective Sizes. TKGEffSize IKGEffSize. Knowledge Contribution (Internet). IKCSize. Discussion. DISCSize DISCDen DISCStr. Note: 1. Code TK and IK represent ICT-practice and ICT-in-Education-practice respectively.. 3.4.1.2. Research Variables The research variables derived from hypotheses 1 to 8 are illustrated in Figure 3.3.. Figure 3.3 Research Variables and Hypotheses of Prestige and Knowledge Exchange Variables ATKCSize and IKCSize represent the degree measurements of knowledge contribution for ICT practice and IK practice respectively. Variables CSize, CDen and CTKCStr: represent variables of personal network size, network density, and tie strength of knowledge exchange, respectively. 34.

參考文獻

相關文件

To investigate the characteristics of Tsongkhapa’s meditation thought, the study is divided into five parts: (1) introduction, (2) Tsongkhapa’s exposition of meditation practice,

 If a DSS school charges a school fee exceeding 2/3 and up to 2 & 1/3 of the DSS unit subsidy rate, then for every additional dollar charged over and above 2/3 of the DSS

FIGURE 23.22 CONTOUR LINES, CURVES OF CONSTANT ELEVATION.. for a uniform field, a point charge, and an

Without using ruler, tearing/cutting of paper or drawing any line, use the square paper provided (Appendix A) to fold the figure with the same conditions as figure 8b, but the area

– A finance charge will be levied if you fail to repay the outstanding balance of retail purchase or cash advances on the payment due date.. 

• A sequence of numbers between 1 and d results in a walk on the graph if given the starting node.. – E.g., (1, 3, 2, 2, 1, 3) from

drives, memory cards, USB flash drives, ExpressCard modules, cloud storage, CDs, DVDs, and Blu-ray Discs, tape, smart cards,. and microfilm and microfiche Internal hard

Figure 6.19 The structure of a class describing a laser weapon in a.