Developing communities of practice model to enhance knowledge and improve learning among faculty members ( Case study : Tabriz University of Medical Sciences )

Introduction Knowledge in medical universities is more important than in other organizations due to the application to health and medical education. On the other hand, considering the complexity of medical education, providing a context for sharing knowledge to overcome these complexities is complicated but equally important. Communities of practice (CoPs) are social structures that focus on knowledge.1 Such communities have been identified as an effective method to extract and disseminate tacit knowledge2 and as a vehicle for learning.3 They are effective tools for the creation and sharing of organizational knowledge, and an increasing number of organizations are adopting them as part of a knowledge management strategy.4 These communities are used as a tool in many higher educational contexts in faculty development 5 and in teacher training.6 CoPs have become commonplace in educational institutions as a means of bringing staff members together to discuss matters of common interest, as well as being used in educational practice itself.7 They can guide the development of interventions to make medical education more effective and can help both learners and educators better cope with medical education complexity.8 Learning is one of the


Introduction
Knowledge in medical universities is more important than in other organizations due to the application to health and medical education.On the other hand, considering the complexity of medical education, providing a context for sharing knowledge to overcome these complexities is complicated but equally important.Communities of practice (CoPs) are social structures that focus on knowledge. 1Such communities have been identified as an effective method to extract and disseminate tacit knowledge 2 and as a vehicle for learning. 3hey are effective tools for the creation and sharing of organizational knowledge, and an increasing number of organizations are adopting them as part of a knowledge management strategy. 4These communities are used as a tool in many higher educational contexts in faculty development 5 and in teacher training. 6CoPs have become commonplace in educational institutions as a means of bringing staff members together to discuss matters of common interest, as well as being used in educational practice itself. 7They can guide the development of interventions to make medical education more effective and can help both learners and educators better cope with medical education complexity. 8Learning is one of the most important benefits of CoPs. 9 CoPs help to improve practice of and learning about teaching. 10CoPs' approach to teaching and learning in higher education provides a space for staff to collaboratively reflect on, review, and update current teaching and learning practices. 11These informal teacher communities allow teachers to engage in a dialogue with colleagues and to share questions, solutions, and interpretations.These teacher communities also provide opportunities to explicate tacit expertise, which helps participants develop an idea of their own role and form a frame of reference for their own experiences. 12The application of CoPs can potentially serve as an effective learning strategy for higher education classrooms by contributing to student professional development while fostering a desire for life-long learning. 13The learning that occurs within CoPs is interactive. 14Interaction between community members increases knowledge, helps access to different experiences, and improves participant's expertise.Improvements in work performance are largely due to the sharing of experiences and best practices. 9Individuals participate in CoPs to learn practical knowledge from one another, and to accomplish shared objectives. 15egarding the responsibility of medical universities in the field of education and health, as well as the diversity of educational trends and complexity of the nature of education in the field of medical sciences, having a systematic approach to health-related topics is necessary.Because it is not possible to achieve all these facets through formal education alone, creating and fostering CoPs at medical universities is an effective way to address these complex needs due to the semi-formal nature of these communities.No studies were done on CoPs at Tabriz University of Medical Sciences (TUOMS); thus researchers conducted this study and presented a native model of CoPs to provide a suitable framework to facilitate knowledge sharing and experience of faculty members at this university under different goals and conditions.

Materials and Methods
This descriptive-survey study was conducted with 245 faculty members at TUOMS in 2017-2018.7][18] and considering a 15% dropout rate.First, the number of faculty members who had the ranks of professor, associate professor or assistant professor was assessed.Then faculty members with administrative positions, members of educational and research councils, and participants in workshops during 2016-2017 were identified, and among them 7 people per item were selected as a statistical sample.Faculty members were selected using simple random sampling.Inclusion criteria were being a faculty member at TUOMS with the ranks of professor, associate professor, or assistant professor with more than one year of activity or instructors who were promoted to assistant professor), or who had participation in two or more workshops during the last two years, and the chancellors and managers of departments, programs, and research centers, supervisors, research directors, educational and research deputies, and research group members.There were no exclusion criteria.The data collection tool was a questionnaire that was valid and reliable in a previous published study.It was designed on a 5-point Likert scale (1= absolutely proper, 2= proper, 3= partly proper, 4= improper, 5= absolutely improper).The impact scores of items were above 1.5, with a content validity ratio (CVR) of 0.78, a content validity index (CVI) of 0.92, an α of 0.89, and an intra-class correlation coefficient (ICC) of 0.92. 19he justification for performing factor analysis was tested using Kaiser-Meyer-Olkin (KMO) and Bartlett's test of sphericity.Mardia's coefficient measurement was used to verify multivariate normality of data.The number of components was determined using exploratory factor analysis (EFA), principal component analysis (PCA) and a scree test that was described with details in a previously published study. 19Confirmatory factor analysis (CFA) was performed in 2 steps to confirm the extracted factors, evaluate the goodness of fit indices and to develop the CoPs model.In the first-order CFA, the relation between each latent variable with observable variables was measured and the partial and general fit indices were calculated.Standard coefficients, t-values, chi-square, root mean square error of approximation (RMSEA), comparative fit index (CFI), normed fit index (NFI), goodness of fit index (GFI), adjusted goodness of fit index (AGFI), and incremental fit index (IFI) were used for determining fit indices.The most important fit index is chi-square, which shows the difference between the observable and estimated matrices.The smaller values of this statistic show the goodness of fit of model, but since this statistic is sensitive to the sample size, it is divided into degrees of freedom in high samples; a value of 2 is often considered appropriate in use of this index, 20 although some sources have suggested that the χ 2 /df should be less than 3 to accept the model. 21The IFI, CFI, NFI, GFI, and AGFI indices are placed between 0 and 1; as these values approach 1, the model is regarded as more appropriate and has goodness of fit. 17,22][25] To investigate the relation between observable and latent variables and the relation between endogenous latent variables and their main constructs (exogenous variables), a second order CFA was done.The reliability of each factor was then calculated using Cronbach alpha.Data were analyzed using SPSS version 21 and LISREL 8.80.

Results
Of 245 faculty members, 230 (94%) returned questionnaires.Of the 230 returned questionnaires, 20 were dropped due to lack of complete answers to all questions for a response rate of 86%.Despite multiple requests, 15 faculty members did not return a questionnaire.Over half of the participants were male (58.1%).The department with the highest representation was Medicine (37%), followed by Pharmacy (13%), and then Research Centers (9%).In terms of rank, the majority (62.9%) were assistant professors (Table 1).The KMO was 0.881 and Bartlett's test of sphericity was performed with a result of 3906.519 and P < 0.001, which justified performing the factor analysis.Mardia's coefficient (2.93) expressed multivariate kurtosis normality. 26Six components were identified based on EFA, PCA and the scree plot (Figure 1 and Table 2 in a previous published study). 19The initial structure of the extracted components was examined in terms of the content relation of items with their underlying factor based on the theoretical principles of the subject.At this stage, four items (14, 22, 25, and  26) were removed due to an insignificant relationship with a factor.The final structure of the extracted components was named according to the latent concepts in items and referring to the related research literature.The results are presented in Table 2. Table 3 shows the partial fit indices (standard coefficients and t-values) for each component.According to Table 2, the results indicate a correlation between observable and latent variables.The relation between each item and its related factor in all items had t-values above 1.96, indicating the significance of the relationships.In Table 4, results of calculating general fit indices of the measurement models by estimating maximum likelihood are presented.Given the comparison of the calculated indices with the acceptable values, the validity of the evaluation scale with the measurement models is able to be confirmed.As already stated, the CoPs model consists of six variables.The standard factor loadings and t-values were calculated to evaluate partial fit indices as presented in Figures 1  and 2. As shown in Figure 1, there is a correlation between observable and latent variables as well as endogenous latent variables and their main constructs (exogenous variables).The strongest correlations were seen between CoPs and informal structures (0.87), interactions (0.83), and participatory leadership (0.82), but formal structure had a weaker relation with CoPs (0.56).The results of examining the significance of the above relations are presented in Figure 2.
In Figure 2, the t-values are above 1.96 in all cases, showing significant relations between the items and their related factors, as well as between the endogenous and exogenous latent variables, which implies the appropriateness of partial fit indices.
Results of determining general fit indices of the second order CFA are presented in Table 5.
Based on results of Table 5, the validity of the CoPs model was confirmed (to achieve the optimal model, several error covariance for the items were released during modification of the model.)The CoPs model following modification is presented in Figure 3.
The results show that after modification of the model, a powerful correlation was found between CoPs and informal structure (0.88), participatory leadership (0.87) and interactions (0.85).
The results of examining the reliability of factors are presented in Table 6.
As shown in Table 6, the results for all variables were higher than the generally accepted value of 0.7, from which it can be concluded that the structures have an optimal level of internal consistency (Figure 4). 27

Discussion
Today's successful organizations must be regarded as institutions where knowledge and skills are continually developed, refined, updated, and protected through complex learning processes that lead to innovation. 14This is possible by providing a suitable context that facilitates the sharing of knowledge and skills among individuals.CoPs are one such structure that enables sharing knowledge and experiences among members of an organization.Regarding attention to the importance of these communities in educational organizations, especially higher education, and taking into consideration that no previous studies were conducted in this field at TUOMS, this study aimed   to identify components of CoPs at TUOMS and to provide a native model.The important point of this study is the presentation of a novel framework.This study identified 6 components of CoPs as well as outlining the structure of CoPs.In theoretical principles and previous studies, selecting the type of structure was based on the approach of community formation by management (a top-down approach) or at the request of the members (a bottom-up approach); the results of this study showed the approach of establishing CoPs relied on many factors, 28 the style of management and leadership, [29][30][31] the type of participation of members (formal or informal), the way members participated (compulsory, voluntary or a combination of both), the member selection process (open or closed), as well as the presence of members (permanent or temporary) in these communities 32,33 ; thus, the formal or informal nature of the structure of these communities can be determined.The results of this study demonstrated the importance of informal structures in CoPs.Another result of this study was the boundary and size of CoPs, which in theoretical principles and, according to previous studies, were examined separately.The boundaries of CoPs are considered as places for negotiation, regeneration of knowledge, 34 promoting learning 35 and are important factors in the innovation process, 35,36 which can be inter-unit, intra-organizational, and interorganizational.Some of these communities may be small and may include only a few experts, while others may consist of hundreds of people. 37In all types of boundaries, people with different levels of expertise and skill through communication and interactions share knowledge and experience inter-or intra-organizationally, which leads to improved education and learning, professional skills development, performance improvement and increased trust.The size of CoPs (number of members) can be determined by the type of objective and the complexity of the issues that need to be discussed, but small As shown in Table 7, the results for all variables were higher than the generally accepted value of 0.7, from which it can be concluded that the structures have an optimal level of internal consistency (Figure 4).

Discussion
Today's successful organizations must be regarded as institutions where knowledge and skills are continually developed, refined, updated, and protected through complex learning processes that lead to innovation. 14This is possible by providing a suitable context that facilitates the sharing of knowledge and skills among individuals.CoPs are one such structure that enables sharing knowledge and experiences among members of an organization.
Regarding attention to the importance of these communities in educational organizations, especially higher education, and taking into consideration that no previous studies were conducted in this field at TUMS, this study aimed to identify components of CoPs at TUMS and to provide a native model.The important point of this study is the presentation of a novel framework.This study identified 6 components of CoPs as well as outlining the structure of CoPs.In theoretical principles and previous studies, selecting the type of structure was based on the approach of community formation by management (a top-down approach) or at the request of the members (a bottom-up approach); the results of this study showed the approach of establishing CoPs relied on many factors, 28 the style of management and leadership, [29][30][31] the type of participation of members (formal or informal), the way members participated (compulsory, voluntary or a combination of both), the member selection process (open or closed), as well as the presence of members (permanent or temporary) in these communities [32][33] ; thus, the formal or informal nature of the structure of these communities can be determined.The results of this study demonstrated the importance of informal structures in CoPs.

Literature review about CoPs
Design and psychometrics CoPs questionnare

Structural equations modeling
communities limit ideas and experiences that are shared to solve problems and achieve other goals.Therefore, according to opinion of faculty members at TUOMS, the presence of more members from within the organization as well as from different units or other organizations is a good approach to achieve goals.
The results of this study regarding interactive processes and participation of CoPs members were congruent with the results of previous research. 32,33All necessary aspects for the interaction of an organization members, a unit of the organization or other organizations were considered, and the unlimited presence of members in CoPs, the participation of members in the activities of CoPs both formally and informally, and their interactions into a combination of face-to-face and virtual communication, increased the desire of members to engage in the activities of the CoPs, consequently, facilitated the creation and sharing of more knowledge and experiences.The presence of an external facilitator helps facilitate the relations among the members of CoPs 7 and may be regarded as a necessary arrangement to achieve the goals.
CoPs are formed for a variety of purposes, including: to enhance problem solving, to transfer best practices and experiences, to provide stewards of knowledge, and to aid in innovation. 37,38The results of this study provided an empirical confirmation of theoretical principles and were congruent with the results of the study of Piri. 39n addition to confirming the role of leadership as one of the most influential factors in the success of CoPs 30,31,37,40 this contention has been supported and confirmed by various researches.In this study, the participatory leadership style, with all its relevant features including trust between members and leader, the commitment of members and the leader, the role of leader or member of the communities as facilitator, was identified and verified as one of the most important components of CoPs.The results were congruent with previous studies about the trust between the leader and the members, 41,42 the commitment of the leader and the members, 43,44 and facilitating role of the leader in CoPs. 40,45

Limitations
No participation of faculty members as instructors leads to non-generalizability of the results to the entire faculty.All participants were faculty members, which limits the external validity and generalizability of the results.Due to the implementation of the study at TUOMS, the results should not be generalized to other universities, but may provide a framework for research at other institutions.It is recommended that CoPs be examined in other studies from the perspective of the faculty members who are instructors and the results be compared with the findings of the present study.It is also suggested that the role of the CoPs model on the performance of faculty members be examined.

Conclusion
Considering the complexity and sensitivity of the duties and functions of medical universities, the wide range of trends and fields of study, as well as the importance of interdisciplinary and multidisciplinary education, research, and learning in these universities, we can use the model presented in this study as a tool for facilitating the creation and sharing of knowledge, promoting learning and innovation, and improving professional skills and performance at the individual, organizational and community levels.Using this model and based on the type of objective, structure, leadership style and management, interactive practices, the level of homogeneity of individuals' specialties, and the number of members, CoPs at the university level can be created and developed.

Table 1 .
Demographic characteristics of participants Figure 1.Standard Coefficients of CoPs Model (CFA2).

Table 2 .
Final structure of the extracted components

Table 3 .
Standard coefficients and t-values of measuring models

Table 5 .
Fit indices of CFA2

Table 6 .
Reliability of Components