4. Results
4.1. Results of Study 1
4.1.6. Social Configurations across School Boundaries
Figure 4.2 graphically depicts knowledge acquisition of ICT practices across school boundaries, drawn with the aid of NetDraw 2.055 software (Borgatti, 2002).
The red nodes indicate the top 5 prestigious STCs. The drawing also presents in blue the STCs of ICT exemplar schools, for reference. The purple nodes indicate that the STCs are prestigious and work in ICT exemplar schools. Figure 4.3 graphically depicts the knowledge acquisition networks of ICT-in-education practices. The red and the purple colored nodes indicate the prestigious STCs in relation to ICT practices.
Top prestigious STCs in relation to ICT-in-education practices are marked by larger circles. It is obvious that the prestigious STCs involved in knowledge sharing of ICT practices are not necessarily prestigious in terms of sharing knowledge of ICT-in-education practices. STCs having positional benefits for their knowledge sharing of ICT practices are not certain to have positional benefits for knowledge sharing of ICT-in-education practices. STC A009 controls the knowledge transmission path (see Figure 4.3), which is peripheral to knowledge sharing with regard to ICT-in-education.
Figure 4.2 Illustration of Technology Knowledge Acquisition Networks
Figure 4.3 Illustration of ICT-in-education Knowledge Acquisition Networks Figure 4.4 graphically depicts the knowledge contribution networks for ICT practices. The colored nodes are provided for reference. Top prestigious STCs in relation to ICT practices are not necessarily perceived as contributing more. STC A006 and A007 are STCs of ICT exemplar schools who contribute less. Figure 4.5 graphically depicts knowledge contribution networks for ICT-in-education practices.
The colored nodes are provided for reference. More prestigious STCs of ICT practices and STCs of ICT exemplar schools are perceived to contribute more than other STCs.
(see Figure 4.4). However, the network illustrates that there is less knowledge sharing of ICT-in-education than of ICT practices.
Figure 4.4 Illustration of ICT Knowledge Contribution Networks
Figure 4.5 Illustration of ICT-in-education Knowledge Contribution Networks
4.1.7. Effects of Structural Holes
Multiple regression equations have been used to test the predictabilities of relational properties with regard to knowledge prestige that STCs acquire for different practices. Note that working in the ICT-exemplar school (variable E-School) has a positive impact on knowledge prestige. Although STCs are the executives of ICT-in-Education exemplary projects, the results show that they did not engage often in knowledge transmission of ICT-in-education practices. The seniority variable (STCAge) has the predictability of prestige in the network of ICT practices.
Evidence (see Table 4.6) shows a small predictability of tie strength to the prestige scores of two networks of practices. The prestigious STCs do not develop stronger tie strengths with other STCs. It conforms (but not significantly) to the arguments that weaker contacts are sources of new information and are important for the dispersion of information (Granovetter 1973; Burt 1992). The predictabilities of network density with regard to the prestige scores of two networks of practices are negative. It verifies the arguments that dense networks are deficient in knowledge prestige. Evidences also show the effective size of knowledge contribution has significant predictability with regard to prestige in two networks of practices. Namely, the prestigious STCs in relation to ICT practices are posited in the strategic position of structural holes and have effective relationships of knowledge sharing. Meanwhile, the prestigious STCs in relation to ICT-in-education practices have the same knowledge advantages. Figure 4.6 and Figure 4.7 illustrate the foregoing graphically.
Table 4.6. Results of the Multiple Regression Analysis for Predicting Prominence
TKprestige TKprestige IKprestige IKprestige IKprestige Variables
NUMofClass -.211 -.077 -.055 .016 .019 STCAge .368** .225** .201 .140* .070 E-School .392** .281** .325* .253** .256*
TKCEffSize .674** .497**
ATKCStr .092 .108 ATKCDen -.067 .115 IKCEffSize .869**
AIKCStr .030 AIKCDen -.173**
R2 0.297 0.776 0.15 0.884 0.493
* p < 0.1
**p < 0.05
Figure 4.6 Result of Hypothesis Testing of ICT Practices
Figure 4.7 Result of Hypothesis Testing of ICT-in-education Practices
4.2. Results of Study 2
4.2.1. Demographic Descriptions
The majority (90%) of the subjects in this study were male. The subjects of this
study had been working as STCs for 1 to 10 years (3.6 years on average). There were 13 schools that had implemented ICT-in-Education exemplary projects.
4.2.2. Relational Properties
The descriptive statistics and Pearson correlations of the relational variables and friendship variable have been assessed in Table D-3 (see Appendix). The personal network sizes and tie strength of all types of practices are highly correlated. That is, STCs develop similar social relationships for each type of knowledge-sharing and friendships across school boundaries. However, the tie strength of friendships is not correlated to the tie strength of knowledge sharing. The correlations of network density show various cohesion interactions for various practices. STCs share knowledge of technology more cohesively than knowledge of digital materials in instructions
4.2.3. QAP Correlations between Different Types of PSNs
Table 4.7 shows the QAP correlations between six matrices. The correlates of five networks of knowledge acquisition are quite similar. The friendship network has a slightly lower association with knowledge acquisition networks.
Table 4.7. QAP Correlations between Five Types of Knowledge Acquiring Networks
HW SW NW PRO DM FR HW
SW 0.81**
NW 0.78** 0.79**
PRO 0.75** 0.75** 0.78**
DM 0.71** 0.74** 0.74** 0.81**
FR 0.65** 0.66** 0.68** 0.67** 0.63**
p** < 0.01
4.2.4. Knowledge Brokerage
The statistical brokerage scores and their correlations among STCs of E-Schools and other schools for five types of knowledge have been assessed in Table D-4 (See
Appendix D). The brokerage scores of friendship are included for comparison. In general, STCs play similar coordinator roles in five types of knowledge sharing. STCs also play similar consultant roles in five types of knowledge sharing. STCs who play representative roles in knowledge sharing of hardware maintenance also are representatives of computer networking knowledge. STCs who are gatekeepers of procurement knowledge also act as gatekeepers of knowledge of instructional digital materials.
The average measurements of four types of brokerage scores for five types of knowledge and friendship are illustrated in Figure 4.8.
Figure 4 ge
4.2.5. Knowledge Brokerage of STCs in E-Schools
ficantly higher consultant scores in a
.8 Average Scores of Knowledge Brokers for Different Types of Knowled
Table 4.8 shows STCs of E-Schools have signi
ll practices. Namely, they have greater opportunities to act as knowledge consultants, especially in regard to the acquisition of digital material skills by STCs of other schools. This also shows STCs have negative scores as knowledge coordinators.
That is, they do not develop highly coordinative relationships with each other.
Especially, they do not exchange computer software utilities. Evidences of Gatekeeper scores have shown that STCs in E-Schools often acquire computer
Table 4.8 Correlations of Four Knowledge Brokerage Scores and STCs in E-Schools hardware and network knowledge from STCs of other schools and contribute them to STCs in other E-Schools.
Coordinator Represent Gatekeeper Consultant HW –0.27 0.31* 0.00 0.50**
SW –0.35* –0.04 0.22 0.39**
NW –0.24 0.37** 0.07 0.33*
PRO –0.11 0.10 0.05 0.43**
DM –0.06 0.01 0.15 0.50**
FR –0.31* 0.17 –0.08 0.39**
p** < 0.0
Table 4.9 Correlations of Brokerage Scores and Friendship for STCs in E-Schools
1 p* < 0.5
HW SW NW PRO DM SW .752
NW .951* .675
PRO .919 .853 .765
DM .742 .894 .542 .942
FR .998** .772 .928 .944 .784
p* 1
p* < 0.5
The evidences presented in Table 4.9 indicate that friendship has a significant impa
* < 0.0
ct on the brokering of hardware knowledge by STCs in E-Schools. In general, the evidences show that friendship has certain correlations with the relationships of brokering computer hardware, network, and procurement knowledge.
5. Discussions
Most of the assumptions of our analytical models have been proven. The results show that knowledge prestige, effective network sizes, and knowledge brokerage are positional advantages that come from knowledge sharing by STCs. Knowledge contribution facilitates knowledge sharing. The following sections will synthesize the findings.
5.1. Knowledge Prestige
Prestige and contribution have been verified as two major factors that drive knowledge exchange among STCs. Ego-centric network surveys led us to investigate network variables of informal knowledge-exchange personalization. An ICT-in-education exemplar school (E-school) indicator was significantly associated with high prestige scores, indicating that high prestige scores were derived for STCs with network locations at higher-status schools. We used this indicator to examine other effective measurements for hypothesis testing, and to discuss knowledge-exchange characteristics.
5.1.1. Characteristics of knowledge contribution
The results verify the assumption that a gift-giving attitude in online communities is related to prestige and can mediate the effects of limited informal personalization to allow for the attainment of knowledge prestige. This may imply that an STC who provides information to online communities can assist large numbers of STCs at other schools and become prestigious. As Walsh and Bayma argue (1996), computer-mediated communication (CMC) provides new opportunities and resources to STCs located at less prominent schools. With the higher accessibility or efficiency of online community technology vis-à-vis knowledge acquisition, STCs might tend to search out supports in an online community when encountering problems; prestigious
STCs may also be perceived as engaging more frequently in online contributions than in directly contributing knowledge to STCs at other schools. That is, they can gain prestige through the Internet. On the other hand, recalling a correlate refers to recalling an impressive relationship (Cross et al., 2001a), which implies that prestigious STCs are not perceived as having disseminated and reciprocated more of their own personal knowledge than they have contributed to STCs at other schools.
The results also indicate that the tie strength with regard to personal knowledge reciprocation is significantly associated with knowledge prestige. That is, prestigious STCs develop stronger reciprocation with advisers. Tie strength in regard to personal reciprocation has a greater effect on prestige than the effects that result from the E-school variable. This implies that the correlation between informal personalization and prestige is stronger than that which exists between authoritative relations and prestige. Because strong ties are recognized to carry information pertaining to the interests of a group to its individual members (Lai & Wong, 2002) and to create solidarity (Burt, 1992), the results indicated that prestigious STCs have strengthened informal personalization leading to concentrated conversations with correlates, and hence, to tacit understanding with correlates. However, the results also indicate that prestigious STCs do not expend as much effort helping STCs at other schools;
indeed, they reciprocate extensively only toward a smaller group from whom they themselves have received advice. This reveals that there is a lack of opportunity for tacit knowledge exchange in a loosely coupled community.
5.1.2. Characteristics and deficiencies of PSNs
The findings reveal that STCs with large knowledge-exchange PSNs have developed wider networks for contributing to STCs at other schools; however, they do not have higher knowledge prestige. Dense knowledge-exchange PSNs are positively associated with online knowledge contributions, reflecting the relative ease of
contributing to an online community rather than to individuals. Tie strength for knowledge reciprocation negatively correlates with online contribution, but not significantly so. Nonetheless, it is obvious that STCs develop weak online relationships. Researchers have indicated that weak ties carry new information among heterogeneous social circles (Granovetter, 1973; Burt, 1992) and facilitate personal status attainment (Lin, 2001). The results imply the opportunities for weak relationships in online knowledge transmission.
It is noteworthy that the variable of tie strength for personal knowledge acquisition is unreliable in the model. This may indicate that STCs who develop stronger ties for knowledge acquisition do not necessarily reciprocate and contribute equally. Frequent reciprocation may depend more on a gift-giving attitude than on a belief in reciprocity.
This study assumed the effects of maintaining PSNs for knowledge exchange.
However, descriptive statistics of network variables reveal that STCs did not have large and dense knowledge-exchange PSNs. Prestigious STCs have small PSNs for knowledge exchange, indicating that PSNs might be less important when more and more information is available through impersonal, online contributions. The most efficient means of exchanging knowledge might be for all to share in a central repository that practitioners would access as needed. However, tacit knowledge requires concentrated interaction and observation, similar to that between an apprentice and a master. The findings of this study imply that online communities do not provide opportunities to facilitate tacit knowledge exchange. The recent advent of instant online communication technology, which facilitates short and frequent interactions, might help to strengthen online relationships and encourage tacit knowledge interaction. However, this assumption needs to be verified in the future.
5.2. Knowledge Sharing across School Boundaries
The results verified that structural-hole of knowledge-contribution across school boundaries has higher effects on knowledge prestige, while network density and tie strength do not. It implies that prestigious STCs contribute knowledge to other STCs efficiently with less redundant ties. The weakly informal personalization of knowledge contribution across school boundaries has positional and informational benefits. However, concentrated conversations leading to a tacit understanding rely on stronger ties. The results indicate that prestigious STCs do not frequently help STCs at other schools even though they are strategically located in the social networks.
It implies that there is a lack of incentive for tacit knowledge sharing in a loosely coupled community.
5.2.1. Configurations of Social Network
Table 4.2 shows the QAP correlations between the six matrices of knowledge sharing across school boundaries. There is less association between network correlates of different types of knowledge sharing. This implies that correlates of network of ICT practices and those of networks of ICT-in-education practices are quite different. Gender is not a significantly influential factor in the selection of correlates. Networks of ICT-in-education practices are more regionally associated than those of ICT practices.
Comparing the results of the Pearson correlation of relational properties and QAP analysis reveals that STCs have similar personal network sizes and tie strength for knowledge acquisition and contribution in both types of practices. That is, STCs having larger networks and stronger contact ties in ICT practices also develop larger networks and stronger contact ties in ICT-in-Education practices. However, they contact different correlates for different types of practices. Researchers have indicated that goal-driven activities are helpful in fostering exchanges of practice
experiences and developing strong relationships (Henri & Pudelko, 2003). The results imply that STCs are aware of the helpers when they seek or contribute help.
The discrepancies between knowledge sharing of ICT-in-education and ICT among STCs are made obvious by contrasting the configurations of social networks across school boundaries. Generally speaking, STCs have small networks of knowledge sharing across school boundaries. The scale of knowledge networks of ICT-in-education is smaller than the scale of knowledge networks of ICT. The scale of knowledge contributions is also smaller than the scale of knowledge acquisition.
There is less association between relational properties of knowledge sharing in ICT and those in ICT-in-education networks. However, STCs having larger network sizes of ICT would have larger network sizes of ICT-in-education. In other words, knowledge sharing of ICT and ICT-in-education across school boundaries is sparser (see Table D-2). Correlates of different knowledge networks have the least associations.
In general, STCs had acquired more assistance in ICT than in ICT-in-education across school boundaries. They did not acquire knowledge of ICT-in-education from acquaintances in their networks where they acquire ICT help. It suggests that STCs of ICT-in-education exemplar schools are not always specialists of ICT-in-education.
STCs of ICT-in-education exemplar schools do not play central roles in knowledge sharing of ICT-in-education across school boundaries.
5.2.2. Deficiency of Knowledge Sharing of ICT-in-education
The results show that there needs to be more connections for knowledge contribution in ICT-in-education. Although much funding has been invested in creating online communities of ICT teachers for this purpose, there are still few weak relationships among STCs across school boundaries. Evidence shows that STCs share few practices of ICT-in-education with each other. This might imply that they are not
entirely engaged in practices of ICT-in-education or they are not responsible for the major work of ICT-in-education on the campuses.
Moreover, the fact of low contact frequencies between prestigious STCs of ICT-in-education and other STCs should be noted. It implies that prestigious STCs have not developed relationships of concentrated discussion with STCs at other schools. This reveals the lack of incentive for tacit knowledge sharing among STCs.
Authorities should reconsider the roles of STCs in long-term strategies to help them meet their responsibilities with regard to ICT-in-education. Some prestigious or central STCs in networks of ICT-in-education (see Figure 4.4 and 4.5), who are not working in the ICT-in-education exemplar schools, have been scouted out in this study. These STCs have positional advantages for transmitting information on ICT-in-education. To facilitate knowledge sharing of ICT-in-education in school contexts, these STCs might serve as new exemplars and good disseminators of knowledge innovations.
5.3. Opportunities of Brokering Knowledge across School Boundaries
Generally speaking, the relational properties of five types of knowledge acquisition across school boundaries have highly significant associations. There is also significant association between correlates of knowledge acquisition networks.This implies that STCs display a consistent perspective on five types of knowledge.
The fact that the tie strength of friendships is not correlated to the tie strength of knowledge acquisition shows that STCs develop higher contact frequencies of knowledge acquisition through professional relationships rather than through social friendships. This might imply that STCs acquire information or explicit knowledge from friendly STCs but develop relationships of tacit understanding with knowledgeable STCs.
The average brokerage scores among STCs of E-Schools and other schools for
five practices are slightly different in degree. STCs’ correlates of friendships are slightly different from their correlates of knowledge acquisition. STCs play similar coordinator roles and consultant roles in knowledge acquisition for five practices.
STCs play a representative role in knowledge acquisition pertaining to hardware maintenance and computer networking on account of their ICT competencies. STCs who act as gatekeepers of procurement knowledge also acquire knowledge of digitalizing instructional materials. This implies STCs in exemplar schools acquire or contribute this specific knowledge from or to STCs in other schools. That is, the exchanges are significant among STCs in ICT-in-education exemplar schools and STCs in other schools.
However, knowledge brokerage scores of STCs in ICT-in-education exemplar schools are more highly correlated to friendships (see Table 4.12.), especially for knowledge of hardware maintenance. This might possibly imply that a clique shares computer hardware, network, and procurement knowledge with correlates on the basis of friendships. On the other hand, it may also imply that they have had opportunities to consult with STCs of other schools on practices of computer software and digital instructional materials, with a view to professional development.
6. Conclusion & Recommendations for Future Studies
This research took a step in exploring the social networks that underpin knowledge sharing in a loosely coupled CoP. Using prestige as an indicator of knowledge sharing, the associations of relational variables and prestige were used to reveal characteristics of knowledge sharing. The findings and the good-fit testing of the path model verify assumptions of knowledge exchange. Online knowledge contributions have mediation effects between limited informal personalization and prestige in a loosely coupled CoP. The findings of the regression and correlation analysis also verify the predictability of social resources for knowledge advantages.
The configurations of knowledge networks across school boundaries reveal the
The configurations of knowledge networks across school boundaries reveal the