A New Framework for Examining Knowledge Management Adoption for Participatory Networks: Crossing All Four Levels of the Digital Divide

Author: Todd Marshall (Syracuse University)

Introduction

The purpose of this paper is to further discussion of the adoption of Participatory Networks as knowledge management systems. The method is conceptual and practical as opposed to empirical. Participatory Networks are selected as an example of a knowledge management system for two reasons: the venue for this paper is KMAfrica, “a continental Participatory Network,” and Participatory Networks are the primary topic of the author’s doctoral research. The author’s definition of a Participatory Network is beyond the scope of this paper, but a brief definition and description are given. The perspective is that a concrete definition of a Participatory Network will provide a context for the examples and provide a baseline for defining Participatory Network as a type of knowledge management system. This will be accomplished by first discussing and defining Participatory Networks. Next, it will trace the development of the Technology Adoption Model (TAM). A description follows this framework, which the author believes moves beyond TAM by incorporating discussions related to adoption in the context of the digital divide, including benefits and potential advantages of this framework. Finally, some practical observations based on the author’s personal experience deploying Participatory Networks are shared.

Participatory Network as a Knowledge Management System

What is a “Participatory Network?” The term “Participatory Network” has been used in many contexts with or without a technological component. Contexts include: democratic movements (Lord & Beetham, 2001; Rakpong, 2004; Shade, 1999), policy (Von Bernstorff, 2003), changing methods of production (Bauwens, 2005; Fonseca, 2004; Uricchio, 2004), libraries (Lankes, Silverstein, Nicholson, & Marshall, 2007), collaborative research networks (Daly, Jogerst, & Schmuch, 2007; Sæther, 2007), and education (Laverack & Dap, 2003; Miles, 1999). A full conceptual exposition of Participatory Networks is beyond the scope of this paper. However, because the literature lacks a clear definition, the author proposes the following definition: “A Participatory Network is an interconnected set of cognizing agents where every member has the potential to converse with other members in a technologically intermediated environment resulting in an entailment mesh to address problems in a given context.” This definition is based on Conversation Theory as developed by Gordon Pask (Pask, 1973, 1975a, 1975b, 1976a, 1976b). At its heart, Conversation Theory is an epistemology about how knowledge is constructed through cyclical and iterative conversations (Pask, 1976a). Conversation and conversing in this context are not simply metaphors and, as Luppicini points out, should not be confused with interaction or communication (Luppicini, 2008). Conversation theory provides two basic frameworks, “a structure for the architecture of conversations” and “a schema for modeling the evolution of conversations” (Pangaro, 2008, p. 36). The definition in simpler terms contains the following elements, 1) a bounded set of individuals (members), 2) a social network, 3) potential to converse (in Pask’s sense of conversation), 4) technical intermediation, 5) an artifact preserving the conversation (entailment mesh as per conversation theory), 6) problems, and 7) a context. Examples of group collaboration and sharing of knowledge could include wikis, social media sites, asynchronous learning networks, group support systems, groupware, and many others.

Information Systems and Adoption and Usage

Knowledge Management Systems are created to be accepted, adopted, and used. However, not all systems are equally adopted or used. Information Systems as a discipline has invested a great deal of effort in attempting to explain the factors that lead to usage and success. Unfortunately, there is no commonly accepted definition of “successful” adoption or usage. A primary weakness in this approach has been a lack of clarity in conceptualizing usage and adoption which has obfuscated research (Johansson & Mollstedt, 2003; Karahanna, Straub, & Chervany, 1999). In this paper, “usage” will be used in the generic sense, a system that is being used, not just initial usage or “adoption,” but to the full spectrum from pre-adoption through post-adoption (Karahanna et al., 1999).

One popular stream of thought in Information Systems concerning adoption is identified by its most popular manifestation, the Technology Acceptance Model (TAM). It began with the Theory of Reasoned Action (TRA) (Ajzen & Fishbein, 1973) and developed into the Theory of Reasoned Behavior (TRB) (Ajzen, 1985, 1991) and Technology Acceptance Model (TAM) (Davis, 1989). While these are not identical, they all center on the ideas that “perceived ease of use” and “perceived usefulness” affect “behavioral intention” resulting in usage. As Wixom and Todd (2005) point out, the focus is on the user’s beliefs and attitudes toward an act leading ultimately to the behavior itself. Despite the longevity of the underlying presuppositions, it is not without its critics and identified weaknesses (Venkatesh, Morris, Davis, & Davis, 2003). While empirical tests of TAM have demonstrated significant results, it has been seen as incomplete and faced constant revisions leading to TAM2 (Venkatesh & Davis, 2000) and eventually TAM3 (Venkatesh & Bala, 2008). At each turn more external factors were identified. TAM2 added “social influence processes (subjective norm, voluntariness, and image) and cognitive instrumental processes (job relevance, output quality, result demonstrability, and perceived ease of use)” (Venkatesh & Davis, 2000, p. 187). TAM3 added more factors (Venkatesh & Bala, 2008).

One of the more recent variants is the Unified Theory of Acceptance and Use of Technology (UTAUT) (Venkatesh et al., 2003). It added three factors affecting “behavioral intention” (performance expectancy, effort expectancy, social influence) and one factor affecting “use behavior” (facilitating conditions). Then, there are the factors which affect the influence of all the factors above (facilitating conditions, gender, age, experience, and voluntariness of use (Venkatesh et al., 2003). While this moves in the right direction, it seems almost like an “everything affects everything” model without much logical or chronological sense. Despite statistical correlation findings, one may ask “How is this really useful?” People follow their intentions and many things affect their intentions. The growing complexity of “everything before intention” seems to indicate a fundamental weakness of this basic approach of the intention based model of usage or “To the man who has a hammer, the whole world becomes a nail.” In this case, “To theories based on TRA, everything affects intention.” It would seem prudent to take a step back and look at those factors which affect intention. UTAUT purports to account for “as much as 70% of the variance in user intention,” however “future research should focus on identifying constructs that can add to the predication of intention and behavior over and above what is already known and understood” (Venkatesh et al., 2003). While this is progress, the solution may require going beyond intention all together. There are occasional insights such as those who would adapt this to look at participation using the same construct (Yoo, Suh, & Lee, 2002), but some might see this as simply replacing the labels on the boxes: “perceived usefulness” with “managing strategy,” “perceived ease of use” with “IS quality,” “intention” with “sense of community,” “attitude” with “visit,” and “usage behavior” with “participation.” Participation is certainly a broader concept than behavioral usage, and community aspects are vital, but the old frame does not have the flexibility that is necessary. Finally, one variant that incorporates social influence (Dholakia, Bagozzi, & Pearo, 2004). This model steps outside of a work/job environment by incorporating self-discovery, social enhancement, and entertainment value and begins to broaden the scope of usage to account for voluntary systems.

While these models are a step in the right direction and attempt to provide a richer picture of factors affecting acceptance, there has been a significant bias in their application. The first problem is the bias toward usage in work contexts. The chart (Table 1) adapted from Zhang reveals this bias (Zhang et al., 2002).

Table 1: Frequency of Contexts (Zhang et al., 2002)

Further bias is toward the individual user (93.8%) as opposed to the group (7.7%) (Zhang & Li, 2005). Areas needing further attention are “Cognitive Beliefs and Behavior, Emotion and Affect, and Trust” (Zhang & Li, 2005).

Second, issues such as mandatory versus voluntary usage are not addressed. Someone could perceive a system as not useful, but use it because subjective norms require usage. In fact, Venkatesh and Davis found this very disconnect in their study. “Subjective norm” had a negative correlation (-0.047, p<.001) with “perceived usefulness” but a positive correlation (0.44, p<.001) with “intention to use” (Venkatesh & Davis, 2000, p. 197).

Third, issues of access, ability, and policy which might hinder participation are not addressed by this stream. An increasing number of Participatory Networks are voluntary and outside the traditional employer/employee context. Usage may require upgrading one’s own hardware, self-training, and use may be a strictly personal choice. TAM does not address barriers to usage.

This author’s opinion is that the over emphasis on “intention” has been a limiting factor toward a better understanding of usage. Others have pointed out the value of the socio-organizational context (Avgerou, 2001). However, when attempting to factor in broader issues such as national culture, they seem to be forced into a TAM framework (Srite & Karahanna, 2006). This has continued to the development of TAM3, expanding TAM from the original 3 factors to 16 where it appears that everything affects everything (Venkatesh & Bala, 2008).

Access and the Digital Divide

In addition to the Information Systems perspective, another stream of academic research emerges from the “digital divide” discussions (Warschauer, 2002). Earlier discussion focused attention on access to broadband as well as socioeconomic and political issues. Over time, this began to include skills most commonly known as information, digital, or technical literacy. Technology and competency are the first and second level digital divides.

The original concept indicated that the primary barrier to adoption was access. Initial digital divide discussions were about whether schools had sufficient access to ICT. It was couched in terms of “haves” and “have nots.” This led to the push for increased access to the internet in public schools, universities, and libraries (Hargitai, 2002; Warschauer, 2002). The assumption was that if people had access, they would take advantage of the opportunity and adopt these new technologies. However, it has been demonstrated that access does not guarantee usage (Crump and McIlroy, 2003). Bridging the access gap still did not create adoption and integration (Crump, 2003). Possession of ICT did not guarantee adoption of ICT.

Following the “bridging” of the first level digital divide, attention shifted to information literacy skills and the educational component of adoption. Access alone was deemed insufficient (Hargittai, 2002; Waschauer, 2002). This conclusion has led to an emphasis on technical competencies and educational initiatives (Dewan and Riggins, 2005). Because of progress in bridging the first level divide over the past decade, ICT access is commonplace in schools, public libraries, and many home settings. Most of the academic work at this time is investigating competencies and education so that students will know how to take full advantage of this technology.

This is basically where the conversation stands at this point. Most studies have been weak on theory and demonstrated little explanatory power (Becker, 1999). I propose that is due to issues related to policy, culture, values, and beliefs which need to be addressed. These factors have not gone unnoticed, but neither have they been integrated into a broader framework, model, or theory (Becker, 2007; Warschauer, 2002; Mardis et al., 2008). Some have pointed out the importance of the political or environmental context (Marcovitz, 2006), but this is not yet widely accepted or identified as a third level (Korupp and Szydlik, 2005). Personal values, a fourth level, have been identified as significant to adoption (Garthwait, 2005; Warschauer, 2002). Other than an earlier paper using the conceptualization in this paper (Mardis et al., 2008), the concept of a fourth level digital divide has not yet emerged. The Behavioral Adoption Framework seeks to integrate first and second level digital divide issues with what it calls third and fourth level divides. The third level consists of cultural values including the school, government, and societal factors that affect adoption. The fourth level is the individual teacher or student’s choice to adopt based on individual, personal values.

Proposed Framework

The theories and models discussed earlier have advanced the study of usage and adoption. However, it is time to reframe the conversation and look at a new framework which accounts for a greater number of factors, but has the flexibility necessary for an increasingly complex contextual environment. In terms of the digital divide issues, this framework would represent a four level digital divide: technology, competency, cultural values, and personal values. In terms of classical Information Systems discussion, this moves past TAM approaches for explaining the dynamics of usage. The proposed framework does not assume that potential users have access or that they have the necessary knowledge, skills, or competencies necessary for adoption. In terms of the digital divide, the factors correspond to the four levels of the digital divide. It may seem obvious that access to technology is required, but it is also obvious that intention to use must precede use, as in TAM. The framework which is proposed here is represented by four factors, each of which is representative of two types of factors. This is the Behavioral Adoption Framework (BAF) which is a revision of an earlier model by the author (Marshall, 2007, 2008) and has been applied to digital library usage in schools (Mardis, Hoffman, & Marshall, 2008). Figure 3 depicts this framework in terms of relationships.

Figure 3

This framework may also be represented as a four quadrant matrix that classifies the factors (Figure 4).
Behavioral Adoption Framework as a Matrix

It may seem obvious that access to technology is required, but it is absent from the TAM models. However, usage in terms of digital divide has focused on the lack of technology or access. In terms of the digital divide, the quadrants represent the four levels of the digital divide: access, education/training, culture/policy, and personal desire. Some authors, such as Korupp and Szydlik (2005) have tentatively identified this third level. However, the concept of a fourth level digital divide is a new idea.

Level 1 – Technology. This is the objective technological artifact and its environment. It is the connectivity, the hardware, software, and the physical setting of usage. This is where the typical issues related to the level one digital divide reside. The issues which affect usage include access to the necessary technology, including internet access, and suitability of hardware and software for the participatory environment. It is technical support and the reliability of systems. As an example, a person may desire to participate in a virtual community such as Second Life, however if they do not have broadband and a sufficiently fast computer, their technological environment prevents their usage. However, if they do have broadband and their computer has insufficient memory and continually crashes during participation, then this will affect their usage. Technology consists of the artifacts and resources necessary for adopting, including the ability to purchase that technology.

Level 2 – Competency. Even if a user has broadband and a sufficiently fast computer, they still must have the knowledge and skills necessary to operate that technology. The focus here is on factors which require the user to interact with the technology. This will vary from user to user even in a single technical environment. For example, in a single department of one company, there will be varying levels of aptitude, knowledge, and experience with a given system. Everyone may have the same technology and same connectivity, but some people are more competent than others. Since this ability varies from person to person, these factors are technical but also very personal. They may vary based on physical ability or disability, education, training, or any other HCI factor. Does the user have the competency to use the technology?

As seen in Figure 3, competency affects usability and motivation. This is based on the concept that competence and the ability to freely engage in an activity are directly related to intrinsic motivation and the self-determination model of motivation (Deci, 1975; Ryan and Deci, 2000; Deci and Ryan, 2000; Vallerand et al., 1997). Individuals who feel competent to perform a certain action (adopt technology) will derive greater pleasure, a full sense of enjoyment, and a feeling of autonomy when they perceive themselves to be competent. This increases motivation. High perceptions of competence are also indicative of higher actual competence (Deci, 1975), thereby increasing ability. As such, the factor of competency incorporates both actual competency and perceived competency.

Level 3 – Socio-Cultural Values. This is the atmosphere surrounding the user. It involves factors related to policy, values, beliefs, culture, willingness, and social influences. One can think of it as the culture in which the user must function. It could be the culture of a school, a business, a home or any other context which affects the user. The context tells the user what behavior is legal or illegal, required or voluntary, encouraged or discouraged, and so on. Whether it is a forum, e-mail, or virtual reality, there are also social norms, policies, and peer pressures which affect usage. The user may or may not choose to follow his or her cultural values but they are a significant factor which does affect the user and his or her motivation.

Level 4 – Personal Values. This is the user’s attitude toward the behavior. Here is intention, desire, pleasure, enjoyment, fulfillment, anxiety and all the other factors which distinguish one user from another and the user from their environment. In short, this is “what’s going on” in the user’s head when she forms opinions and ultimately decides whether she will or will not use a given technology. Classical Information System theories of usage and participation have focused on this quadrant because this is the location of the factors, including intention, which are seen to lead to the final decision regarding behavior.

Some may view this quadrant as the most significant because they perceive ultimate decisions to adopt as happening here. However, there are situations when other factors outweigh personal values. For example, full intention to use and high motivation are always limited by the technology, the context, and competency. Factors in quadrant number four may lead the individual to change their technology, learn new skills, or change their environment. However, such a dynamic is not directly leading to adoption but to changing oneself or one’s environment. So, this issue is not directly addressed by this framework.
Finally, personal values should not be confused with personal characteristics such as demographics. Studies continually compare usage to age, income, race, education, geography, and similar factors (Lenhart, 2002). This framework does not ignore those factors, but instead understands them to influence each of the four quadrants. Race and education will certainly affect one’s cultural and personal values and the technology at one’s disposal. However, such factors do not cause usage. Demographics alone are not deterministic of usage. Rather, they shape an individual’s values and abilities. From the standpoint of this framework such factors are secondary.

The purpose of these quadrants is not to identify every possible factor, but to propose a taxonomy which could be both explanatory and predictive. In this sense, it can serve as a taxonomy of factors, aid in discovery, and serve as a basis for further discussions. Specifics in quadrant each can and will change based on the context, but the basic issues should remain the same. The framework also has philosophical symmetry, addressing issues which are technical and social as well as issues which are internal and external. This may provide face validity, but the framework still must be tested to be proven accurate. The following is a proposed formula for BAF if one were to be able to create a quantifiable scale for each factor.

Technology * Competency = Ability
Cultural Values * Personal Values * Competency = Motivation
Motivation * Values = Usage

This is similar to Vroom’s Expectancy Theory: Expectancy * Instrumentality * Valence = Motivation (Vroom, 1964). Effects of weak or null values demonstrate that all factors must be necessary for usage. If one of the four factors is weak, usage will be significantly affected.

Finally, there are several assumptions which are implied in this framework. First, genuine adoption requires both ability and motivation. Second, all four factors must cooperate at some level for usage to occur. Technical factors are required for the ability to use and social factors in combination with competency are required for motivation. Third, the contents of each factor can change from context to context. This provides flexibility for broad application regardless of the technology or the environment.

Advantages of the Behavioral Adoption Framework

There are at least six benefits to this framework: 1) it moves beyond the construct of user intention (TAM), 2) it provides a robust schema for digital divide discussions, 3) it can encompass factors leading to usage as well as non-usage, 4) it is scalable for simple and complex usage, 5) it addresses both the ability and motivation, 6) it acknowledges the role of cultural influences while still allowing for personal choice.

First, TAM has put “blinders” on research into usage and has created the “illusion of cumulative tradition” when, in fact, the same construct has simply been tested repeatedly with slight variations (Benbasat & Barki, 2007). While this conclusion may seem to make the point of this question and discussion of TAM moot, this is also a case in point that Information Systems needs to return to its roots, work on theoretical conceptions about people, systems, and context, and not be distracted by contextually specific solutions to complex problems. In this situation, Benbasat and Barki suggest moving beyond the current state of affairs: re-examine the original theory, re-conceptualize usage to broader contexts, look at acceptance/adoption/usage as a linear concept, investigate sources of belief, and consider other models which are not based solely on belief (Benbasat & Barki, 2007). BAF moves in this direction.

BAF, as a framework, acknowledges that all four digital divides must be addressed for each user in each usage context. For example, a child may desire to engage in a social network site at home and in free time at school. In this case three factors remain the same. However, the third level (socio-cultural values) may be dramatically different. At school the site is forbidden but at home the site is encouraged. In this case, intention may be the same but values of the school and fear of “getting caught” are the primary factors affecting usage.

Not all usage is “intentional.” In fact, a user may have no intention but use only as a result of the social pressure of the environment. Or, there may be intention but no competency. This is often seen when a person desires to do something such as engage in a collaborative workspace, and even though levels 1, 3, and 4 are “crossed” the person’s skills eventually discourage further participation. BAF acknowledges that even good intentions can fail.

Regardless of the activity, the individual or group of individuals must choose. As an individual must cross all four divides to adoption and usage, so must a group cross the divides. Members of Participatory Networks who are unable to cross all four will not really be members of the Participatory Network. For example, a group support software application can’t be completely successful unless the whole group of users crosses all the divides together. Each individual must all cross and the group as a whole must cross together. In this way, this framework can stimulate discussion not just about why a single individual uses or not, but also why group usage is or is not successful.

Ability and motivation are never completely static. Levels of skill and competency can certainly be reached through practice, but sometimes they can slip. Motivation is even more variable. The initial interaction and excitement of a participatory environment may wane over time as the excitement wears off or the conversation turns dull. BAF does not treat either of these as inherently unchanging constants.

Finally, BAF acknowledges that everyone is affected by culture but still retains volitional power. For example, the author knows an engineer at a large IT company who has adopted and uses the latest home video technologies. However, he ignores the insistence of his colleagues, friends, and families and refuses to join or participate in social networking activities. While it is true that he lives in an individualistic American culture, BAT can explain his activity by recognizing the difference between level 3 and level 4. In contrast a more collective culture may see that the collective pressure of the group is stronger than the desire of the individual and the individual submits to the “will of the group.” The difference between the individualistic engineer who says “no” and the collectivist who says “yes” stems from the differences in the personal values of the two individuals, not just the existence or lack of external pressure. One values the group above self. The other values self above the group.

Practical Implications for Adoption

Through 10 years of attempting to build and deploy Participatory Networks, the author feels that he is finally beginning to ask the right questions rather than having the right answers. The development of BAF and research into the definition of Participatory Networks have not provided answers in and of themselves, but have already been useful in pointing out areas for investigation that had been overlooked. The following discussion provides some of the author’s personal opinions based on first hand situations trying to cross each of the four divides.

The technical divide is the often the easiest divide to cross. While it may seem, and actually is, insurmountable in some contexts, access should be viewed as only the first step in a long journey. It is the beginning, not the end. You might be able to set up a system in a day that someone cannot learn how to use in a year, at least not learn on their own. It’s much easier to install a learning management system than get IT staff, faculty, students, and administration to use it or even read the help section. It is somewhat like giving birth. It’s a painful and messy process that may take years to achieve, but it is just the beginning.

Build into the user’s competencies. Real education requires more than a manual, it requires mentoring. While it can be very difficult to design or adapt a technology to a user, that frustration is less than the frustration of the user giving up because they don’t have the skills and thus loose motivation. If it’s doubtful that the users will be able to acquire the necessary competencies, a great opportunity arises to look for more “low tech” solutions. The author recently tried to convince a school that a high quality “free” video conferencing solution was a poor choice for student interaction because even a major New York University had difficulty dealing with such a complex system. The better and cheaper solution would have been notebooks with free video conferencing software and built in webcams.

Capture the culture. The meaning of “capture” here is to understand culture and control it. IT adoption in any form is as much a cultural process as a technical process. It doesn’t take long to realize that simply giving people access to technology and training them means that they will use it. In one school, the administration was excited to receive wireless hardware and was happy to have it installed. However, after the donor saw it working and departed, it was immediately removed and put back in the box because of concerns over security. The students and teachers were not very happy. In this case, the culture of those with the power was neither understood nor influenced. If it had been understood initially, then the possibility for influence could have been examined. In this case the power of the greatest cultural influence, in that context the rector, made all user intention pointless.

Don’t forget the “wetware.” Many would even say it is more about the “wetware” (software between the ears) than about the hardware or the software. No matter how good the technology, the individual user must ultimately make the choice themselves. They must login, answer their messages, or move the mouse. It is very rare to get 100% usage of any technology, especially one with a strong component like a Participatory Network. Sometimes putting 20% of one’s energy into forcing the final 3% to participate just isn’t worth it. One must also remember that every culture has people who are socially challenged or people who prefer to watch instead of participate. Sharing knowledge or not sharing can also be a power issue as opposed to a technical issue. Maybe the personal participation in the network is more about the individual’s personal gain or loss than what is best for the group.

Conclusion

As the diversity surrounding adoption and usage of Participatory Networks expands, newer and richer frameworks are needed. With the increasing ubiquity of Participatory Networks outside traditional contexts comes the need for approaches which address the complexity of the user and his or her context. Evaluating usage and adoption of smart phone users logging into Participatory Networks in an African village is obviously more complicated that looking at PC users in a controlled office environment. BAF is an attempt to provide a frame work that can be time sensitive to the user and their context. While it has the weakness of not being tested, the author hopes that this will stimulate further conversation and move the conversation about digital divides beyond issues of technology and competency.

Bibliography

1. Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. Action-control: From cognition to behavior, 11, 39.

2. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211.
3. Ajzen, I., & Fishbein, M. (1973). Attitudinal and normative variables as predictors of specific behaviors. Journal of Personality and Social Psychology, 27(1), 41-57.
4. Avgerou, C. (2001). The significance of context in information systems and organizational change. Information Systems Journal, 11(1), 43-63.
5. Bauwens, M. (2005). The political economy of peer production. 1000 Days of Theory, 1.
6. Benbasat, I., & Barki, H. (2007). Quo vadis, TAM? Journal of the Association for Information Systems, 8(4), 211-218.
7. Daly, J. M., Jogerst, G. J., & Schmuch, G. A. (2007). APS Participatory Network Case Study Review. SOCIAL WORK IN HEALTH CARE, 46(1), 21.
8. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319-340.
9. Dholakia, U. M., Bagozzi, R. P., & Pearo, L. K. (2004). A social influence model of consumer participation in network-and small-group-based virtual communities. International Journal of Research in Marketing, 21(3), 241-263.
10. Fonseca, M. F. (2004). Alternative certification and a network conformity assessment approach. The Organic Standard, Torfolk, GROLING AB, 4-6.
11. Johansson, N., & Mollstedt, U. (2003). A conceptual model to assess and evaluate use of a new ICT-based service. Paper presented at the ECITE 2003, Madrid Spain, September 25-26.
12. Karahanna, E., Straub, D., & Chervany, N. (1999). Information Technology Adoption Across Time: A Cross-Sectional Comparison of Pre-Adoption and Post-Adoption Beliefs. Management Information Systems Quarterly, 23(1), 14.
13. Lankes, R. D., Silverstein, J., Nicholson, S., & Marshall, T. E. (2007). Participatory Networks: The Library as Conversation. INFORMATION TECHNOLOGY AND LIBRARIES, 26(4), 17.
14. Laverack, G., & Dap, D. H. (2003). Transforming information, education and communication in Vietnam. Health Education, 103(6), 363-369.
15. Lord, C., & Beetham, D. (2001). Legitimizing the EU: Is there a'Post-parliamentary Basis' for its Legitimation? Journal of Common Market Studies, 39(3), 443-462.
16. Luppicini, R. (2008). Introducing Conversation Design. In R. Luppicini (Ed.), Handbook of Conversation Design for Instructional Applications (pp. 1-18). Hershey, PA: Information Science Reference.
17. Mardis, M. A., Hoffman, E. S., & Marshall, T. E. (2008). A new framework for understanding educational digital library use: re-examining digital divides in US schools. International Journal on Digital Libraries, 9(1), 19-27.
18. Marshall, T. E. (2007). A New Model for Explaining Technology Usage: The Quadratic Usage Model. Syracuse University Center for Digital Literacy.
19. Marshall, T. E. (2008). A new approach to understanding the digital divide: Why two levels are not enough. Conversants, 1(1).
20. Miles, S. (1999). Creating conversations: The evolution of the Enabling Education Network. In E. Stone (Ed.), Disability and development (pp. 74–88). Leeds: The Disability Press.
21. Pangaro, P. (2008). Instruction for Design and Desings for Conversation. In R. Luppicini (Ed.), Handbook of Conversation
22. Design for Instructional Applications (pp. 35-48). Hershey, PA: Information Science Reference.

23. Pask, G. (1973). A Theory of Conversations and Individuals (Exemplified by the Learning Processes on CASTE).
24. International Journal of Man-Machine Studies, 5(4), 443-566.

25. Pask, G. (1975a). Conversation, Cognition and Learning: A Cybernetic Theory and Methodology: Elsevier Publishing Company.
26. Pask, G. (1975b). The Cybernetics of Human Learning and Performance: A Guide to Theory and Research: Hutchinson Educational.
27. Pask, G. (1976a). Conversation Theory: Applications in Education and Epistemology: Elsevier.
28. Pask, G. (1976b). Styles and strategies of learning. Br. J. educ. Psyschology, 46, 128-148.
29. Rakpong, P. (2004). The Promotion of A Participatory Network and Power Decentralization Case Study: The Upper Mae
30. Ping Basin in Chiangdao District of Chiang Mai Province. Bangkok: United Nations Development Program (UNDP).

31. Sæther, B. (2007). From Researching Regions at a Distance to Participatory Network Building: Integrating Action Research and Economic Geography. Systemic Practice and Action Research, 20(1), 15-25.
32. Shade, L. R. (1999). Roughing It in the Electronic Bush: Community Networking in Canada. Canadian Journal of Communication, 24(2).
33. Srite, M., & Karahanna, E. (2006). The Role of Espoused National Cultural Values in Technology Acceptance. Management Information Systems Quarterly, 30(3), 679.
34. Uricchio, W. (2004). Beyond the Great Divide: Collaborative Networks and the Challenge to Dominant Conceptions of Creative Industries. International Journal of Cultural Studies, 7(1), 79.
35. Venkatesh, V., & Bala, H. (2008). Technology Acceptance Model 3 and a Research Agenda on Interventions. Decision Sciences, 39(2), 273-315.
36. Venkatesh, V., & Davis, F. D. (2000). A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies. MANAGEMENT SCIENCE, 46(2), 186-204.
37. Venkatesh, V., Morris, M., Davis, G. B., & Davis, F. D. (2003). User acceptance of informations technology: Toward a unified view. MIS Quarterly, 27(3), 425-478.
38. Von Bernstorff, J. (2003). Democratic Global Internet Regulation? Governance Networks, International Law and the Shadow of Hegemony. European Law Journal, 9(4), 511-526.
39. Wixom, B. H., & Todd, P. A. (2005). A Theoretical Integration of User Satisfaction and Technology Acceptance. Information Systems Research, 16(1), 85.
40. Yoo, W. S., Suh, K. S., & Lee, M. B. (2002). Exploring the Factors Enhancing Member Participation in Virtual Communities. Journal of Global Information Management, 10(3), 55-71.
41. Zhang, P., Benbasat, I., Carey, J., Davis, F., Galletta, D., & Strong, D. (2002). Human-Computer Interaction Research in the MIS Discipline. Communications of the AIS, 9(20), 334-355.
42. Zhang, P., & Li, N. (2005). The Intellectual Development of Human-Computer Interaction Research: A Critical Assessment of the MIS Literature (1990-2002). Journal of the Association for Information Systems, 6(11), 227-292.