This paper investigates the factors affecting medical bloggers' knowledge sharing behavior from both personal and external aspects. We develop a model based on the social cognitive theory and augment it with the social capital theory. The model is empirically examined based on the survey data collected from 75 bloggers writing on medical issues, and evaluated with confirmatory factor analysis and structural equation modeling. Among the personal factors, we examined outcome expectations including reputation and enjoyment in helping others. Encouragement by others, identification and interaction ties were investigated as external factors. Enjoyment in helping others and reputation were found to have significant direct affect on medical bloggers' knowledge sharing behavior, while encouragement by others, identification, and interaction ties showed no significant direct affect. However, encouragement by others exhibited significant impact on reputation and enjoyment in helping others. The implications for theory and practice, and future possible research are discussed.
The growing importance of knowledge as a sustainable competitive resource has compelled many organizations to launch knowledge management initiatives and technologies. However, organizational knowledge assets growth is contingent upon the willingness of individuals to share their experiences, insights, and wisdom with the others in inter/intra organizations (Davenport & Prusak, 1998; Jarvenpaa & Staples, 2000; Kim & Mauborgne, 1998). The fact is that the availability of information systems such as intranets, distributed libraries, document management systems, and groupware applications is no guarantee that knowledge sharing will actually take place (Alavi & Leidner, 2001; Jarvenpaa & Staples, 2000; Orlikowski, 1996).
Besides, hoarding knowledge and reluctance to share it are natural human tendencies (Davenport & Prusak, 1998). This implies that making individuals' knowledge into organizational knowledge is a challenging issue (Grant, 1996; Nonaka & Takeuchi, 1995; Ryu, Ho, & Han, 2003). Thus, it is critical to understand factors that support or constraint individuals' knowledge sharing behavior.
Healthcare organizations such as hospitals are one of the typical knowledge intensive organizations. Therefore, knowledge management is also a crucial process in this type of settings. It can help to optimize the support and primary health processes, increase the quality of healthcare, efficiencies and effectiveness, and improve learning capacity of the organizations (Mansingh, Osei-Bryson, & Reichgelt, 2009; Wahle & Groothuis, 2005). In healthcare knowledge management, similar to knowledge management in other settings, knowledge sharing between different actors has been identified as a critical factor to achieve knowledge management benefits (Mansingh, Osei-Bryson, & Reichgelt, 2009; Richardson & Committee on Quality of Health Care in America, 2001). Moreover, in medicine and healthcare organizations losing the opportunity of having knowledge may lead to a medical error. In other words, knowledge sharing may no longer be a “nice to have” process; but changes into a “must have” one (Naeema, 2007).
Theoretical and practical medical knowledge is vital to the healthcare of patients. Medical knowledge sharing is also essential for healthcare professionals such as nurses or physicians. This is in accordance with an ongoing need for elevating the quality and efficiency of the care for patients. Healthcare professionals “are required to be research-oriented, creative in medical care, and ready to take new medical knowledge opportunities that can be acquired through various organizational mechanisms” (Ryu, Ho, & Han, 2003). Thus, healthcare workers who have acquired unique specialized knowledge have to share their knowledgefor patient care improvement (Mansingh, Osei-Bryson, & Reichgelt, 2009).
There are various researches on knowledge management in healthcare organizations. Nevertheless, only a few of them theoretically study knowledge sharing (Bo & Cong-dong, 2005). The essential role of medical knowledge and medical knowledge sharing in providing high quality healthcare for patients and few cases of relevant theoretical studies on this topic provide the motivation for this research to investigate medical knowledge sharing.
Among new online communication technologies, blogs or weblogs are becoming increasingly popular. As of April 2006, Technorati (a blog search engine) was tracking over 70 million blogs, and in September 2008, it was indexing 133 million blogs. Blogs are personal frequently updated websites with latest entries (posts) on top of the previous content (Bhatt, 2005). In addition to posts, blogs usually contain web links to other blogs and web sites. Another important part of blogs called “comments”, which is attached to each entry, allows blog readers to leave their comments for each post in an interactive format.
The content of blogs and their topic depends on their author's opinion. Blogs may be just personal diaries of bloggers. They may also provide commentary or news on a particular topic. The spread of information is very fast on blogs, even faster than mainstream media. Especially during and after the 2003 Iraq war, blogs have been increasingly noticed due to their power in providing breaking news and different viewpoints. However, the content of blogs is not restricted to this type of information. Bloggers also publish their own knowledge, experience and thoughts on their blog.
Thus, blogging as a pervasive online activity facilitates communicating ideas, dispersing information, and sharing knowledge to a large number of users in a short time. “Blogging impacts not only the mainstream media and private internet users but also business, to the point that blogs can no longer be ignored by organizations. Well known companies such as Microsoft, Sun Microsystems, and Google use them internally as knowledge sharing and communication tools as well as externally to express their point of view to the public” (Habermann, 2005, p. 35).
While one of the challenging issues in knowledge management is the individuals' natural tendency to hoard knowledge (Bock, Zmud, & Kim, 2005; Davenport & Prusak, 1998), bloggers voluntarily and frequently publish their knowledge of different subjects on their blogs. In spite of the vast knowledge sharing activities in blogosphere, there are few but growing studies on knowledge sharing behavior in this interesting context. This offers another motivation for the present research to explore knowledge sharing behavior in the blogosphere setting.
However, various contents are available in blogosphere, and it is reasonable to expect disagreement on perceiving all these contents as knowledge. Thus, it is rational to know about categorizations of blogs especially those classifications that are based on the content of blogs. In one classification, four basic types are considered for blogs according to two dimensions: personal vs. topical, and individual vs. community (Krishnamurthy, 2002). Another classification which is based on the content of blogs is: filter blog (“primarily containing observations and evaluations of external, typically public events”), personal journal blog (“primarily reporting events in the blogger's life and the blogger's internal states and/or reflections”), and k-log which stands for knowledge blog (“primarily containing information and observations focused around a (n external) topic, project or product”) (Herring, et al., 2005, p.147). In this study, the focus is on the topical, individual blogs according to Krishnamurthy's classification. According to Herring's classification we deal with k-logs. The special topic in the blogs we examine is medical issues. We call them as medical blogs. These blogs are mainly written by physicians, nurses, residents or MD students.
The present study aims to investigate medical knowledge sharing behavior in the online context of blogosphere. The objective is to understand the factors that promote or constraint medical bloggers' knowledge sharing behavior. In order to acquiring better insight into such factors, the study utilizes the social cognitive theory and augments it with the social capital theory. With the application of the social capital and the social cognitive perspectives, the factors will be explored from both external and personal standpoints. Thus, the specific research question of this study is: “What are the factors that affect knowledge sharing behavior of medical bloggers?”
The remainder of this study is organized as follows: First, the theoretical background of the study will be presented. Then, a model developed based on the two theories and proposed hypotheses for examining the research question will be provided. Next, the research methodology and data analysis will be explained. Finally, implications, limitations, and possible future research opportunities will be discussed.
Theoretical Background and Research Model
The extant literature on knowledge sharing behavior has determined various factors that influence individuals' tendency to contribute knowledge inside or outside organizations via various technologies (e.g. electronic knowledge repositories, electronic networks of practice). Such factors are personal motivations (Wasko & Faraj, 2005), costs and benefits (Kankanhalli, Tan, & Wei, 2005), outcome expectations (Chiu, Hsu, & Wang, 2006; Hsu, et al., 2007), social identity (Zheng & Kim, 2007), contextual factors and organizational climate (Bock, Zmud, & Kim, 2005; Kankanhalli, Tan, & Wei, 2005 ). One possible categorization of the factors affecting individuals' knowledge sharing behavior can be personal factors and external influences. While personal factors are those related to individual characteristics in doing the behavior, external ones are imposed factors by the knowledge sharing context or people in that context.
To support this idea and to explore medical bloggers' knowledge sharing behavior from both personal and external aspects, we draw on the social cognitive theory. The social cognitive theory (Bandura, 1977, 1986) is a widely applied theory with demonstrated validity for modeling individuals' behavior in information systems literature (Compeau, & Higgins, 1995a; Liaw, et al., 2006). According to the social cognitive theory, human behavior, environmental influence, cognitive and other personal factors are affected by each other and there exists a triadic reciprocal relationship among them (Bandura, 1986; Wood & Bandura, 1989).
In this study, the focus is on the role of personal factors and external influences on individuals' behavior. The social cognitive theory posits that self-efficacy and outcome expectations are two major predictors of a person's cognition in undertaking a behavior (Bandura, 1986, 1997). Self-efficacy is defined as “People's judgments of their capabilities to organize and execute courses of action required to attain designated types of performances.” (Bandura, 1986, p. 391) and outcome expectation is “a judgment of the likely consequence such performance will produce” (Bandura, 1997, p.21). The effects of self-efficacy has been explored in different IS phenomena such as computer and IT usage (Compeau & Higgins, 1995a; Easley, Devaraj, & Crant, 2003), internet use (Hsu & Chiu 2004; Lam & Lee 2005), use of web based systems (Nahl, 1996; Vijayasarathy, 2004), and it has demonstrated explanatory power in these areas. It has also been employed in knowledge management research to see its effect on knowledge contribution (Bock, & Kim, 2002; Hsu, et al., 2007; Kankanhalli, Tan, & Wei, 2005). We will not consider self-efficacy in this study because of this comprehensive body of knowledge and research on this determinant. Moreover, medical blogging in this study is a voluntary action. If medical bloggers were not confident in their knowledge sharing ability they were unlikely to perform the behavior (Chiu, Hsu, & Wang, 2006). Therefore, one part of the research model in this study explores the relationship between personal cognition (i.e., outcome expectation) and medical bloggers' knowledge sharing behavior.
Social persuasion which is one of the key sources of self-efficacy in the social cognitive theory (Bandura, 1986) can be considered as an external factor. “People are lead, through suggestion, into believing that they can cope successfully with what has overwhelmed them in the past” (Bandura, 1977). The effect of social persuasion was empirically tested in computer use (Compeau & Higgins, 1995b). We also examine the role of social persuasion on knowledge sharing behavior as an external influence.
To examine external factors that influence individuals' behavior, we adopt the social capital theory. The social capital theory is based on the proposition that social relationships constitute productive resources that stimulates social actions for mutual benefits (Coleman, 1988; Putnam, 1995). The social capital concept has been applied in different context to explain various social phenomena (e.g., collective action, community involvement) (Coleman, 1990). There are also some studies that have applied this theory as a framework for understanding knowledge creation and sharing in the context of organizations as well as virtual communities (Chiu, Hsu, & Wang, 2006; Nahapiet & Ghoshal, 1998; Wasko & Faraj, 2005).
There is some debate whether social capital will develop among internet users or not (Putnam, 2000; Uslaner, 2000; Wellman, et al., 2001). However, medical bloggers differ from general internet users, because most of them have shared interest, interdependence, and frequent interaction. All these are requirements of social capital development in a collective (Nahapiet & Ghoshal, 1998). Thus, another part of our research model attempts to address the relationship between external influences with medical bloggers' knowledge sharing behavior by means of social capital and social persuasion concepts. Figure 1 shows the research model of this study.
According to the social cognitive theory, outcome expectations refer to one's beliefs in possible consequences of his/her behavior (Bandura, 1997; Compeau & Higgins, 1995b). Based on the social cognitive theory, people are more likely to undertake a behavior if they anticipate it will lead to positive outcome. In addition to the social capital theory, the social exchange theory also posits that individuals' engagement in a behavior depends on whether its benefits outweigh its costs or not (Bock, Zmud, & Kim, 2005). The positive consequences can be seen as motivations, and human behavior can be controlled by their effects. Thus, the expectation of personal benefits can encourage individuals to share their knowledge with others (Constant, Sproull, & Kiesler, 1996). These benefits can take two forms, extrinsic or intrinsic (Kankanhalli, Tan, & Wei, 2005). Several researches have demonstrated that extrinsic and intrinsic benefits play an important role in knowledge sharing behavior of individuals (Constant, Sproull, & Kiesler, 1996; Kankanhalli, Tan, & Wei, 2005; Wasko & Faraj, 2005). In this study, we explore the effect of reputation (extrinsic benefit) and enjoyment in helping others (intrinsic benefits) on knowledge sharing behavior of medical bloggers.
One potential way an individual can benefit from his/her knowledge sharing can be through enhancement of his/her image in a collective (Wasko & Faraj, 2005). Knowledge contributors can show others that they posses valuable knowledge (Kankanhalli, Tan, & Wei, 2005). This leads to respect and higher reputation for them (Constant, Kiesler, & Sproull, 1994; Constant, Sproull, & Kiesler, 1996). Prior research has found that individuals share their valuable expertise in organizations to achieve better image and prestige (O'Dell & Grayson, 1998; Kankanhalli, Tan, & Wei, 2005; Kollok, 1999). Research in electronic networks also demonstrates that building reputation and gaining status is a strong motivator for active participation (Constant, Sproull, & Kiesler, 1996; Wasko & Faraj, 2005). In addition, an individual's reputation in online settings may transfer to his/her profession (Stewart, 2003). Thus, the perception that sharing knowledge will enhance one's image and status may motivate individuals to share their valuable, personal knowledge with others on their blogs. This leads to our first hypothesis:
H1: Reputation is positively related to medical bloggers' knowledge sharing behavior.
Enjoyment in Helping Others
In addition to extrinsic benefits like improving their image, individuals may also expect intrinsic benefits from their knowledge sharing. One of these intrinsic benefits can be enjoyment in helping others. This benefit is actually based on the concept of altruism. By altruism, one seeks to increase welfare of others, at some cost to him/her (Hars & Ou, 2002). Medical bloggers also share their knowledge on their blogs at some costs (e.g. spending time and effort) to them. Thus, medical bloggers may be motivated to share knowledge due to their tendency to help others and behaving altruistically. Prior research in electronic networks (Wasko & Faraj, 2005), electronic repositories (Kankanhalli, Tan, & Wei, 2005), and open-source programming (Hars & Ou, 2002) also shows that individuals are motivated to contribute their knowledge because of the satisfaction they gain by helping others. This leads to our next hypothesis:
H2: Enjoyment in helping others is positively related to medical bloggers' knowledge sharing behavior.
Social Persuasion and Encouragement
“Social persuasion is a way of increasing people's beliefs that they possess the capability to achieve what they seek” (Wood & Bandura, 1989, p.365). Social persuasion also called verbal persuasion is a form of subjective norms (Compeau & Higgins, 1995b). Subjective norms are crucial in shaping human behaviors. If people receive realistic encouragement, they will be more likely to exert greater effort and become successful than if they are troubled by self-doubt (Bandura, 1986). The morecredible the source of verbal persuasion is for an individual, the more its effect will be on the individuals' behavior (Bandura, 1977, 1986). This leads to the following hypothesis:
H3: Encouragement by others is positively related to medical bloggers' knowledge sharing behavior.
Encouragement also influences outcome expectation. If credible people encourage an individual to accomplish an action, the perceived consequences of that behavior will be affected (Compeau & Higgins, 1995b). Thus, our next hypotheses are:
H4: Encouragement by others is positively related to reputation expectation.
H5: Encouragement by other is positively related to enjoyment in helping others.
To examine the role of external influences on knowledge sharing behavior of medical bloggers, we also draw upon the social capital theory. Social capital is “the sum of the actual and potential resources embedded within, available through, and derived from the network of relationships possessed by an individual or social unit. Social capital thus comprises both the network and the assets that may be mobilized through that network” (Nahapiet & Ghoshal, 1998, p.243). Nahapiet and Ghoshal identify three dimensions for the social capital. 1- Structural capital which is related to the structural links and connections between individuals. 2- Relational capital which is related to the nature of the relationship among the individuals. 3- Cognitive capital which is related to cognitive capability of individuals to understand the knowledge. In this section of our research model, we discuss the role of relational and structural dimensions.
The first dimension of social capital discussed in this study is ‘relational capital’ which refers to the nature of the relationships within a collective (Nahapiet & Ghoshal, 1998). Existence of identification within the collective is prerequisite to creation of relational capital. “Identification is the process whereby individuals see themselves as one with another person or group of people” (Nahapiet & Ghoshal, 1998, p. 256). This process takes place through membership in that group or the act of the group as a reference one (Nahapiet & Ghoshal, 1998). In this study, the group of people is medical bloggers community and identification is an individual's sense of togetherness and positive attitude toward the network of medical blogger. The perception of identity and belonging to a unified group will affect the members' willingness toward following the norms of the group. Specific to this study, it will affect medial bloggers' tendency to share their knowledge on their blogs. This leads to the following hypothesis:
H6: Identification is positively related to medical bloggers' knowledge sharing behavior.
Another dimension of social capital is ‘structural capital’ which refers to the overall connection between members of the collective that is created through social interactions (Burt, 1992; Nahapiet & Ghoshal, 1998). More direct and regular interaction and contact among more members of the collective increases the structural capital of the community. The more rich a community is in terms of structural capital, the more likely it is to observe cooperation and collective action of its members. In the case of knowledge sharing, more social capital of a collective increases the intensity and frequency of knowledge sharing (Chiu, Hsu, & Wang, 2006). In this study, interaction ties are defined in terms of frequent communication with other medical bloggers and time spent on interaction with other medical bloggers. Thus, the last hypothesis is:
H7: Interaction ties are positively related to medical bloggers' knowledge sharing behavior.
The survey methodology was used to collect data for testing the research model of this study, since it increases the generalizability of the results (Dooley, 2001).
Table 1 provides formal definition of the constructs. Whenever possible, items for measuring the constructs were adopted from prior research. The scale items for reputations and enjoyment in helping others were adopted from Kankanhalli, Tan, and Wei (2005) with the wording slightly modified to fit the purpose of this study. Encouragement by others items were adopted from Compeau and Higgins (1995b). Interaction ties and identification items were developed based on Nahapiet and Ghoshal (1998). Knowledge sharing behavior items were adopted from Hsu et al. (2007). The questionnaire contains 22 self-reported questions related to the six research constructs. They were measured using seven-point scales anchored from “strongly agree (1)” to “strongly disagree (7)”. Two more questions were also asked to identify age, gender (demographic variables) of the respondents. To ensure the conceptual validity of the items, one round of unstructured sorting was conducted by four postgraduate IS students. The results showed the need for a few refinements in some of the items wording in order to represent the intended constructs properly. The results of convergent and discriminant validity of the items is explained in data analysis section.
Table 1. Formal Definition of Constructs
Medical bloggers are the target population of this study. Medical bloggers are those bloggers who write on medical issues on their blogs. Searching internet for finding a list of medical blogs resulted in the following website: “http://www.medlogs.com”. It is a medical news and weblog aggregator. Its list of physicians, nurses, residents and students' blogs was used in this study. After finding the blog addresses, each blog was checked to find the blogger's email address. Some of the link lists of these blogs were also used to find more medical bloggers. After retrieving all the email addresses an invitation email was sent to the medical bloggers for participating in a web-based survey. Participants were assured of the confidentiality and academic use of their responses. Web-based survey was conducted by this study because the target subjects were available through internet. The “www.surveymonkey.com” which is an online web survey provider was used for collecting the responses. Three follow up emails also was sent to increase the responses. 89 responses were collected. The exclusion of 14 incomplete responses resulted in a total of 75 valid ones for data analysis. Most of the respondents were female (53.62%) and in the age group of 20 to 30 years (48.48%). Table 2 shows demographic information of respondents.
Table 2. Demographics
Data Analysis and Results
A two-step structural equation modeling analysis (Anderson & Gerbin, 1988) was conducted to analyze the collected data. The first step of the analysis deals with the measurement model and assessments of the instruments reliability and validity. The second step involves the analysis of relationships among latent constructs. Confirmatory factor analysis (CFA) is the conventional statistical method to ensure construct validity. It pre-identifies construct-item correspondences but it allows free changes of factor loadings. Fixed construct-item correspondences are then examined for confirmation.
To satisfy convergent validity, the construct-item correlation should be significant (Anderson & Gerbin, 1988). For an item, the average variance extracted (AVE) by the latent factor should be greater than 0.5. It implies that a construct should explain more than 50% of the item variance (Fornell & Larcker, 1981). Furthermore, items corresponding to same construct should be highly correlated. These correlations are tested using two measures, composite factor reliability (CFR) and Cronbach's alphas (α). They are required to be greater than 0.7 for high correlation indication (Hair, et al., 1995). Table 3 reports the result of convergent validity for our sample using LISREL. All criteria were satisfied.
Table 3. The Convergent Validity of Measurement Model
The discriminant validity of the scales was assessed using Fornell & Larcker's (1981) guideline. According to this guideline inter-construct correlations should be less than the square root of the AVE. It means that an item should be better explained by its corresponding construct than by other constructs. Table 4 lists the correlations among the constructs, with the square root of the AVE on the diagonal. All the diagonal values exceed the inter-construct correlations. Hence, the test of discriminant validity was acceptable.
Table 4. Descriptive Statistics and Factor Correlation
For a measurement model to have a good model fit, the chi-square value normalized by degrees of freedom must be less than 3, the root mean square of error approximation must be less than 0.10, and Non-Normed Fit Index (NNFI) and Comparative Fit Index(CFI) should exceed.90 (Hair, et al., 1995). These model fit indices derived during CFA are provided in Table 5. Our results indicated satisfactory model fit.
Table 5. Model Fit Indices
In summary, our measurement model confirmed discriminant and convergent validity of all factors. Thus we can proceed to next step and test the overall model.
The measurement model exhibited satisfactory model fit. Hence, we proceeded to hypothesis testing. Hypothesis testing was carried out by creating a structural equation model in LISREL. Following structural equation modeling procedure, we arrived at the results summarized in Figure 2.
Before drawing conclusions for hypothesis testing, the model must fit the data well. The model fit indices shown in Table 5 are in acceptable thresholds and largely the same as model fit indices derived from CFA.
The model in Figure 2 explains 63% of the variance in medical bloggers' knowledge sharing behavior, 73% of the variance in reputation expectation, and 86% of the variance in enjoyment in helping others expectation.
Reputation was found to have a significant positive relation (2.13, p<.05) with knowledge sharing behavior. Enjoyment in helping others also exhibited a strong positive effect (3.22, p<.01) on knowledge sharing behavior. Moreover, enjoyment in helping others was the most influential factor in explaining the variance in knowledge sharing behavior. However, encouragement by other showed no significant direct influence (0.01, p=.99) on knowledge sharing behavior, while it exhibited a strong effect on reputation expectation. It was also found to have a strong effect on enjoyment from helping others. In addition, identification was found to have a negative but insignificant effect (−.10, p=.92) on knowledge sharing behavior. Interaction ties also exhibited no significant effect (1.74, p=.08) on knowledge sharing behavior. In summary, four out of seven hypotheses were supported.
The aim of this study was to understand personal and external factors influencing medical bloggers knowledge sharing behavior. Among the personal factors, we examined outcome expectations including reputation and enjoyment in helping others. Encouragement by others, identification and interaction ties were investigated as external factors.
Enjoyment in helping others was found to be the most important factor among all the other factors inevtigated in this study. Reputation was the next influential factor. These results are similar to previous knowledge sharing studies' findings in other contexts (Kankanhalli, Tan, & Wei, 2005; Wasko & Faraj, 2005) which indicates the effect of extrinsic and intrinsic outcome expectations on medical knowledge sharing behavior.
Among external factors, encouragement by others exhibited no significant direct impact on knowledge sharing behavior, but it showed a strong significant effect on reputation and enjoyment in helping others. Therefore, insignificant direct effect of encouragement by others may be due to fully mediation of reputation and enjoyment in helping others.
Contrary to our expectation, identification showed a negative but insignificant relation with knowledge sharing behavior. Interaction ties effect on knowledge sharing behavior was also not significant at p=0.05 level, but significant at p=0.10. One possible explanation may be due to high correlation (r =.65) of identification with interaction ties.
In summary, comparing external factors and personal factors results, personal factors seem to have more effect on medical knowledge sharing behavior than external factors. This trend can be adequately explained with the possibility of low social capital formation among medical bloggers which is similar to previous studies findings in Internet context (Putnam, 2000; Usnaler, 2000).
This study has several theoretical and practical implications. From theoretical perspective, first, it offers a more holistic view of the factors affecting knowledge sharing behavior by utilizing social cognitive theory and augmenting it with social capital theory for explaining external influences. Second, through the lens of these two theories and holistic model, this study gives insight into both personal and external factors that are predicted to affect this behavior. Third, it gives insights into medical knowledge sharing that is a critical and less studied type of knowledge sharing for knowledge management in healthcare sector. Forth it brings social-psychological theoretical explanation into this process in healthcare knowledge management. Fifth, it deals with blogosphere, a new pervasive online communication technology which is experiencing growing interest in academic settings.
From practical point of view, the findings of this study shows blog service providers the importance of specific features that best support use of blogs generally for knowledge sharing and specifically for medical knowledge sharing. According to the results of the study, outcome expectation and social persuasion are important determinants of medical knowledge sharing behavior. Thus, blog service providers may invest on developing relevant features that leverages these expectations of bloggers. Moreover, the findings also offers helpful insight for healthcare knowledge management in relevant organization such as hospital for exploiting blogs as a system for medical knowledge sharing among their employees (e.g., nurses, physicians, surgeons). In other words, it facilitates estimating the advantages, and challenges of using blogs as organizational knowledge management tools.
Limitations and Further Research
The subjects of this study were mostly personal bloggers writing on a specific topic. Therefore, the findings of this study may not be generalizable to organizational blogging purposes. Further research will be required to generalize the model. This study investigated only the role of bloggers and not the readers of the blogs in the social network of blogosphere. There are many blog readers who do not have any blogs by themselves, but they interact with bloggers through leaving comments for them. Future research can investigate the role of blog readers on social network of blogosphere.