Knee Arthroplasty

Permanent Link: http://ufdc.ufl.edu/UFE0042740/00001

Material Information

Title: Knee Arthroplasty Shared Experience in a Virtual Community
Physical Description: 1 online resource (248 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011


Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: Knee Arthroplasty: Shared Experience in a Virtual Community Information once obtained secondhand from family, friends and acquaintances?the two-step-flow of communication?can now easily be found via the Internet. This paper examines how opinion leadership, trust, credibility, and virtual community intersect in a confidential online health community formed around potential, pending, and completed knee replacement. Knee replacement surgery (knee arthroplasty) is the most commonly performed joint replacement procedure. This study utilized Social Network Analysis (SNA), an online survey, and interviews to understand how members of one such online community used information they gathered there. Findings indicate that experiential proximity?impressions that the ?other? has been through similar circumstances??generates perceptions of opinion leadership, trust, and credibility. Results show that how people use online health forums and what it is that attracts people to them is related to how effective forums are in helping users cope with a particular health condition. Peer-based information exchange is valued because of the experiential information available through relationships formed online.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by MICHAEL D BROOKSHIRE.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Ostroff, David H.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042740:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042740/00001

Material Information

Title: Knee Arthroplasty Shared Experience in a Virtual Community
Physical Description: 1 online resource (248 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011


Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: Knee Arthroplasty: Shared Experience in a Virtual Community Information once obtained secondhand from family, friends and acquaintances?the two-step-flow of communication?can now easily be found via the Internet. This paper examines how opinion leadership, trust, credibility, and virtual community intersect in a confidential online health community formed around potential, pending, and completed knee replacement. Knee replacement surgery (knee arthroplasty) is the most commonly performed joint replacement procedure. This study utilized Social Network Analysis (SNA), an online survey, and interviews to understand how members of one such online community used information they gathered there. Findings indicate that experiential proximity?impressions that the ?other? has been through similar circumstances??generates perceptions of opinion leadership, trust, and credibility. Results show that how people use online health forums and what it is that attracts people to them is related to how effective forums are in helping users cope with a particular health condition. Peer-based information exchange is valued because of the experiential information available through relationships formed online.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by MICHAEL D BROOKSHIRE.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Ostroff, David H.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042740:00001

This item has the following downloads:

Full Text




2 2011 Michael D. Brookshire


3 ACKNOWLEDGMENTS First, I would like to thank Burlingame Industries of Rialto, California for its financial support during the time it took to complete this docto ral degree. I will forever be grateful to Robert C. Burlingame, not only for his encouragement, but also for allowing me to work an abbreviated schedule throughout coursework and the writing of this dissertation, creating a working environment that helped alleviate many of the day to day pressures graduate students experience in pursuit of a doctorate. I would like to thank my committee, Chair Dr. David H. Ostroff, members Dr. Cory Armstrong and Dr. Amy Jo Coffey, and outside member Dr. Ken neth Sassaman. My committee made this a rigorous process, and I am the better for that. And a special nod to Dr. Sassaman for offering advice and support beyond this dissertation during a particularly difficult time in the process. A heartfelt thanks also goes out to Ass ociate Dean Dr. Debbie Treise, who provided unwavering support; academic when needed, emotional when the demons of doubt slipped their leash, threatening to render impotent the entire endeavor. Thank yous are in order for Craig Lee, without whom the surv ey component of this paper would never have seen the light of day; Christina Locke, whose formatting skills helped shaped the final product; Paul Kirk, whose help in locating sources proved invaluable to this research, while his humorously intelligent jab s at my theoretical musings slapped me back into reality when I took the process too seriously; Kim Holloway, whose help with, and advice on, all things financial in my time at Florida were immeasurable; and Jody Hedge, to whom I will be eternally grateful for guiding me


4 A very special thanks to Patrick Reakes Chair of the Humanities and Social Sciences Library at Library West whose skillful advice in helping me to find even the most ob scure information in support of this research was incalculable. Just as important, he and his family were friends to me as well. Thanks to Richard Warner, Jeremy Reither, and Jim Lawrence of BoneSmart who trusted this research and provided me access to the BoneSmart community. And I owe a very special thanks to Josephine Fox, whose knowledge of and insight into all things BoneSmart made this research a reality without her this paper would not have been possible. And lastly, I thank my daughter Al granddaughters Mackenzie and Megan, and grandson Sean, for keeping me grounded in reality and connected to my roots, and, when I needed it most, providing safe haven 3,000 miles away from the rigors of academia. And for just being alive.


5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Purpose of Study ................................ ................................ ................................ .... 16 Theoretical Perspectives ................................ ................................ ......................... 17 Methodology ................................ ................................ ................................ ........... 18 Need for Study ................................ ................................ ................................ ........ 19 Contributions to Research ................................ ................................ ...................... 20 Background ................................ ................................ ................................ ............. 21 The Internet and Information Seeking ................................ .............................. 21 Seeking Health Information Online ................................ ................................ ... 25 Joint Replacement ................................ ................................ ............................ 30 BoneSmart.org ................................ ................................ ................................ 33 Further Chapters ................................ ................................ ................................ ..... 35 2 REVIEW OF THE LITERATURE ................................ ................................ ............ 37 Strength of Weak Ties ................................ ................................ ............................ 37 Virtual Community ................................ ................................ ................................ ... 42 Community ................................ ................................ ................................ ....... 43 Commonality ................................ ................................ ................................ .... 45 Identity ................................ ................................ ................................ .............. 47 Space and Place ................................ ................................ .............................. 49 Social Support ................................ ................................ ................................ .. 54 Online Health Community ................................ ................................ ................. 55 Opinion Leadership ................................ ................................ ................................ 56 The Primary Group ................................ ................................ ........................... 59 Social Power ................................ ................................ ................................ .... 61 Defining Opinion Leadership ................................ ................................ ............ 62 Opinion Leadership in a New Age ................................ ................................ .... 64 Social Network Analysis ................................ ................................ ................... 68 Social Network Analysis in Mass Communication ................................ ............ 72 Two Step to Multi Step ................................ ................................ ..................... 76 Antecedents of Effective Communication ................................ ............................... 79 Trust ................................ ................................ ................................ ................. 80


6 Credibility ................................ ................................ ................................ .......... 89 Homophily ................................ ................................ ................................ ........ 93 Gaps in Research ................................ ................................ ................................ ... 96 The Nature of Inquiry ................................ ................................ .............................. 97 Research Questions ................................ ................................ ............................... 99 3 METHODOLOGY ................................ ................................ ................................ .. 102 Mixed Methods ................................ ................................ ................................ ..... 102 Case Study ................................ ................................ ................................ ........... 103 Social Network Analysis as Method ................................ ................................ ...... 104 Online Surveys ................................ ................................ ................................ ..... 110 Online Interviews ................................ ................................ ................................ .. 113 Phenomenology ................................ ................................ ................................ .... 117 Textual Analysis ................................ ................................ ................................ .... 117 Reliability and Validity ................................ ................................ ........................... 118 Definitions ................................ ................................ ................................ ............. 120 Nominal Definitions ................................ ................................ ........................ 121 Operational Definitions ................................ ................................ ................... 121 Sampling and Recruitment ................................ ................................ .................... 122 Procedures ................................ ................................ ................................ ........... 124 SNA Observation ................................ ................................ ............................ 124 Online Questionnaire ................................ ................................ ...................... 128 Online Interviews ................................ ................................ ............................ 129 Research Questions Revisited ................................ ................................ .............. 130 4 FINDINGS ................................ ................................ ................................ ............. 134 SNA Observation ................................ ................................ ................................ .. 135 Online Questionnaire ................................ ................................ ............................ 138 Online Interviews ................................ ................................ ................................ .. 142 Textual Analysis ................................ ................................ ................................ .... 144 Research Questions Resolved ................................ ................................ ............. 146 5 SUMMARY, CONCLUSIONS, DISCUSSION, AND RECOMMENDATIONS ....... 166 Summary ................................ ................................ ................................ .............. 166 Conclusions ................................ ................................ ................................ .......... 171 Discussion ................................ ................................ ................................ ............ 174 Further Theoretical Implications ................................ ................................ ............ 187 Limitations ................................ ................................ ................................ ............. 189 Recommendations ................................ ................................ ................................ 192 APPENDIX A ONLINE QUESTIONNAIRE ................................ ................................ .................. 197 B INTRODUCTORY E MAILS ................................ ................................ .................. 205


7 C ONLINE INTERVIEW QUESTIONS ................................ ................................ ...... 206 D ONLINE INTERVIEW RESPONSES ................................ ................................ .... 209 LIST OF REFERENCES ................................ ................................ ............................. 222 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 247


8 LIST OF TABLES Table page 1 1 Percentage of internet users who l ook online for specific health topics .............. 36 3 1 Operational definitions of major concepts ................................ ......................... 132 3 2 Example of SNA matrix ................................ ................................ .................... 132 4 1 Op Forum (partial) .................. 160 4 2 rtial) ............... 162 4 3 Characteristics of survey participants ................................ ............................... 162 4 4 Nominations for leaders in BoneSmart Knee Forum ................................ ...... 163 4 5 Trust by length of membership ................................ ................................ ......... 164 4 6 Descriptive statistics for credibility measures ................................ ................... 164 4 7 Means table: Homophily by education ................................ .............................. 165


9 LIST OF FIGURES Figure page 1 1 Percentage of American adult internet users who say th ey belong to this type online group. ................................ ................................ ................................ ....... 36 3 1 Sociogram of Knee Pre Op Forum (partial) ................................ ...................... 133 4 1 Sociogram of Pre Op Forum members ................................ ............................. 159 4 2 Sociogram of Recovery Forum members ................................ ......................... 161 5 1 A: Original two step flow of communication. B: Multi step flow of influential communication in an online knee health community. ................................ ....... 196


10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Deg ree of Doctor of Philosophy KNEE ARTHROPLASTY: SHARED EXPERIENCE IN A VIRTUAL COMMUNITY By Michael D. Brookshire May 2011 Chair: David H. Ostroff Major: Mass Communication There is no argument that the internet has become fundamental to living in conte mporary society. Information once obtained secondhand from family, friends and acquaintances the two step flow of communication can now easily be found via the internet Social science research has indicated that in any community there are individuals to w hom others look for opinions and advice on various issues. This paper examines how opinion leadership, trust, credibility, and virtual community intersect in a confidential online hea lth community formed around potenti al, pending, and completed knee repla cement Knee replacement surgery, or knee arthroplasty, is the most common joint replacement procedure performed today. Strength of weak ties also informs this research as information deemed more valuable to community members regarding knee replacement cam e from associations formed online rather than from This mixed method case study utilized Social Network Analysis (SNA), an online survey, and phenomenological interviews to understand how members of one such online community used information they gathered there. A mixed methods approach also helped extract accounts related to how being a member of this community shaped


11 of it self. Findings indicate that experiential proximity been through similar circumstances generates perceptions of opinion leadership, trust, and credibility. Results indicate that how people use online health forums and what it is that attracts people to them is related to how effective forums are in helping users cope with a particular health condition. This study extends previous research by further demonstrating that peer based information exchange is valued because of the experiential information available through relationships formed online. This inquiry contributes to current literature on health related online forums by examining in more detail what motivat es people to make use of these forums and linking this empirical ly to how effective people perceive these forums to be.


12 CHAPTER 1 INTRODUCTION Computer mediated communication, of course, is not just a tool; it is at once technology, medium, and engine of social relations. It not only structures social relations, it is the space within which relations occur and the tool that individuals use to enter that space. It is more than the context in which social relations occur (although it is that too), for it is commented on and imaginatively constructed by symbolic proces ses initiated and maintained by individuals and groups (S. G. Jones, 1995, p. 16). In a world where information is freely available across multiple platforms, people are developing migratory patterns of information seeking behavior as they are able to go a lmost a nywhere, via technology, in the search for knowledge Users must active ly participate as they ar e motivated in some cases forced by necessity to seek information and make connections across varying media systems, competing media economies, divergen t cultures, and national borders. More than 1.5 bil lion people worldwide now use the internet, both at work and in their social lives (Internet World Stats, 2010). It has changed the way people inform, amuse, educate and care for themselves, and the ways they pray, bank, work, shop and stay in touch with others (Pew internet & American Life Project, 2008). Over the past four de cades, the internet h as grown from an experimental point to point military network to a global platform for technological innovatio n (Rogers, 2001) and, via both wired and wireless technologies, a comprehensive point of departure for access to an incomprehensible amount of information. To some this convergence of technology, industry, and culture leads to social change driven not by m edia appliances, but by social interaction among participants (Jenkins, 2006). An individual living in modern society uses information extracted from media content to, at least in part, help them make sense of their everyday lives. As


13 more information exis ts on a given topic than any one individual can retain, people are inclined to communicate through various media to exchange and accumulate information. Thus, information consumption has become a collective process; no one person can know everything, every individual knows something, and people are now capitalizing on new technologies as they pool their knowledge, skills and resources. The result is a form of collective intelligence that people may utilize in the practice of everyday life and that Jenkins s ees as an alternative to traditional sources of power. When these forms of collective intelligence develop through social interaction via the virtual communities (Rheingold, 1993). This would indicate p e ople are constru cting and spending considerable time in they are occupying cyber landscapes whose construction is achieved through social action (Carter, 2004). As a result, place nitions The virtual community has become one such place where people actively seek social relationships, and these communal associations offer opportunity to create, negotiate and reproduce co mplex social places. Study of this particular type of place hel ps illustrate how people a re extending themselves into cyberspace and how certain aspects of relationships constructed therein are being moved offline and inco rporated into daily life. Although connection to others through the internet is key to virtual c ommunity, the notion of a community is no t applicable to all sites of on line discourse (Ridings & Gefen, 2004). Some discussion groups and chat rooms, for example, are just places for people to meet without any sense of permanence or consistency a mong the members. C hat rooms for example, largely cater to peo ple looking to discuss practically any topic with


14 others each room having a different mix of people each day, none returning on a regular basis. As s uch and because they lack a regular basis of partic ipation by patrons, chat rooms would no t qualify as a virtual communities for many researchers (e.g., Blanchard, 2008 ; D. B oyd & Ellison, 2008 ; Q. Jones, 1997 ). Ridings Gefen, and Arinze (2002 ) offer a comprehensive definition of the term and which this research will follow Virtual communities are groups of people with common interests and practices that [ sic ] communicate regularly and for some duration in an organized way over the internet through a common location or mechanism (p. 273). D ecades of social science research have indicated that in any community there is an individual (or small group of individuals) to whom others look to help them form opinions on various issues (Katz & Laza rsfeld, 1955; King & Summers, 1970; Lazarsfeld, Berelson, & Gaudet, 1948; Rogers, 1962; Rogers & Cartano, 1962; Shah & Scheufele, 2006; Weimann, 1994) T he diffusion of the internet has radically expanded readily available sources for information of all ty pes. Informatio n that was once obtained second hand from friends and acquaintances (i.e., the traditional two step flow hypothesis), is now easily found via the internet E vidence exists to support the contention that the two step flow is mutating into a m ulti step flow of communication ( Case, Johnson, Andrews, Allard, & Kelly, 2004; Rains, 2007; Weimann, 1994 ; Weimann, Tustin, van Vuuren, & Joubert, 2007 ) especially as it relates to health related information. As an aging U.S. population looks for soluti ons to increasing health concerns, it also appears that people are looking to the internet for answers to important


15 health questions to both documentary sources and advice from other laypersons and professionals in discussion forums (Pew internet & America n Life Project 2008). S Fox and Jones (2009) found that 83% of internet users, or 61% of U.S. adults, have looked online for information about a specific disease, a certain treatment, alternative medicine, health insurance, doctors, hospitals, and ways to stay healthy. According to these researchers, of all U.S. adults who looked for health related information online in 2008, 57% said their most recent search had an impact, either major or minor. Of that percentage, 28% said the internet helped connect t hem to expert or professional services, further underscorin g the role the internet plays in linking people to other people in the course of decision making ; 60% said information found online affected a decision about how to treat an illness or condition; 5 6% said it changed their overall approach to maintaining their health or the health of someone they help take care of; 53% said it led them to ask a doctor new questions, or get a second opinion from another doctor; 49% said it changed the way they think about diet, exercise, or stress management; 38% said it affected a decision about whether to see a doctor; 38% said it changed the way they cope with a chronic condition or manage pain (S. Fox & Jones, 2009). Gone it seems, are the days when patients simp ly made the o bvious (and indeed perhaps only) choice of consulting their doctor. The internet and the world wide web have changed the n ature of information seeking. With 74.1% of the United States population identified as users (Internet World Stats, 2010 ), the internet is often the first choice for learning about an array of topics including health topics as well as a medium for discussing those topics with other users (Case et al., 2004) ; to some degree the internet substitutes for older media (Kaye


16 & Jo hnson, 2003). As a result, the emergence of the internet as an omnibus source may also have changed the nature of opinion leadership; both more authoritative (e.g., medical journals) and more interpersonal (e.g., condition specific support groups) sources are readily available and accessible online. The frequ ent conclusion of past research, that most people turn to others (family members, friends, neighbors and nearby expert s, such as physicians) for advice, may be in doubt Consistent with the concept of s ocial community, virtual communities have been i dentified as facilitator s to participating in common activities and sharing emotions and knowledge in an electronic environment (Preece, 2000). M embers who voluntarily participate in these communities are ini tially strangers, but common goals and shared underst andings gradually encourage members to share their knowledge with, provide assistance to, and support each other (Williams, 2006). Critics contend that these types of associations lack traditionally unde rstood commitment and responsibility, therefore levels of intimacy and connection are insufficient for community, especially when participants can log off, shut down or walk away when they so choose (S. G. Jones, 1995). Others however, contend that engagin g and cooperating online in shared practic es and information exchange enhances community by fostering credibility, trust, belo nging, loyalty, mutual 1993; Young & Tseng, 2008). As trust and credibi lity are crucial to both opinion leadership and virtual community, these concepts were investigated as components of this study. Purpose of Study This research aimed to understand how opinion leadership, trust, and credibility operate in a health related online community. It explored how these concepts develop


17 and are maintained in a virtual en vironment, how participants experience membership in the community, and how they experience using the internet and world wide web to help them understand their condi tion. In addition, this study was designed to elicit discussions of how being a member of this community BoneSmart.org, formed around potential, pending, and performed total knee replacement perception of the condition, and ultimately, how they experience the condition in and of itself. Theoretical P erspectives This study is primarily informed by two theoretical perspectives: the two step flow of communication, still considered viable after 70 years of social science research and the inevit able introduction and proliferation of digital communications technologies; and, phenomenology, both a philosophical theory and a methodology, focusing on lived experience. Two step flow attends to the significance of interpersonal processes in the flow of information, and through explanation, description and interpretation, phenomenological inquiry examines how people structure experiences of everyday phenomena (in this case a health related condition) and articulate what these experiences mean from a firs t person perspective (in this case expressed through written communication). strength of weak ties This theory proposes that strangers could offer an advan tage over tho in obtaining useful infor mation. Granovetter argues that strong tie relationships occur among people who are similar in many respects; similar people are likely to know the same things and are unlikely to know dissimilar things. When information is unavailable through strong ties, people may


18 obtain it through weak tie relationships those characterized by absent or infrequent cont act, lack of emotional bonds and no history of reciprocity Weak ties are useful, Granovette r contends, because they are more numerous thereby increasing the probability of obtaining useful information. Numerous rep lies increase the range of counsel, and if the request is multifaceted numerous replies could increase the to tal usefulness of cont ributions (Constant, Sproull, & Kiesler, 1996). Methodology This research utilized a mixed methods approach to ans wer the research questions Fundamentally a case study, this investigation examined in depth a singular social phenomenon : the encompassing experiences of members of a virtual community in which these members gather information and make decisions regarding knee replacement (for themselves or others). After collecting selected postings on the BoneSmart.org website (the subject of this investiga tion), Social Network Analysis (Wellman, 1983; Scott, 2000) was employed to observe if certain individuals were central to communication processes in the community. Using sociometric methods (Katz & Lazarsfeld 1955) to, in part, identify opinion leaders i n the community, an online questionnaire was disseminated which also include d measures of tru st, credibility and homophily. In addition, r esults f rom the questionnaire revealed a wide array of knee health experiences as they related to demographic variabl es, which aided in the selection of interview respondents who represent ed the widest possible range of conditions and circumstances in the community. Systematic analysis of online phenomenological interview s with consenting respondents then extracted the i ndividual, interpersonal and cultural logics that these respondents employed in their communication processes within this community.


19 Need for Study Opinion leadership has largely been researched within the context of traditional forms of community and soc ial interaction. These views hold that individuals sharing geographical space engage in primarily face to face communication discussing politics, consumer products, popular culture, or behaviors (Katz & Lazarsfeld, 1955). Individuals may be influenced by o pinion leaders, but the emergenc e and overwhelming popularity of the internet may be changing t he nature of opinion leadership. D ifferences between online and offline communication are important because assigning trust and credibility in electronic communi ties must accordingly be made primarily on the basis of written communication In virtual setting s the sense of social distance and the lack of social cues make it difficult for people to identify with each other and to assess mutual ability, integrity, a nd benevolence (Young & Tseng, 2008). technological environment may indicate a multi step flow, rather than the traditional two step flow. More research is needed to understand n ot only how opinion leadership operates in online environments, but how trust and credibility are assigned to other people in these environments. The credibility of support providers with in online communities largely derives from experiential proximity or the perception that the provider of support has been throug h similar circumstances, similar problems has engaged in similar behaviors, and has similar attitudes and beliefs abou t the condition he or she faces (Jin, Bloch, & Cameron, 2002; Tanis, 2008 ; Wright, 2000). This study attempted to determine whether that held for this virtual community. Past research has addressed trust and credibility directed at websites or online groups; however, research


20 has been insufficient with regard to how these conce pts develop when the intended object is perceived to be another person Contribution s to R esearch T he contemporary media environment is a constantly changing environment featuring co ntinuous audience fragmentation and affecting ho w individuals gather and act on informatio n in the processes of daily life The evolution of media has altered the logic of social order by restructuring the relationship between physical place and social plac e, changing the ways in which people transmit and receive information P eople depend less today on traditional forms of communication and information gathering television, radio, print, expert professionals, family and friends and more on the internet and world wide web to inform their lives (Estabrook, Witt & Rainie, 2007) This research adds to communications literature by demonstrating how emphasis has shifted from simply documenting the existence of virtual communities to illuminating how larger structural patterns (i.e., rapidly changing technology, access to the interne t, patient/doctor relationships, instant availability of wide ranging sources of health related information that are less accessible offline) affect the ways in which community structures shape social production and reproduction. While each online communit y represents a unique set of technical features or a distinct user culture, examining leadership with in them will offer greater insight into understanding user behavior, thus strengthen ing a general theory of online leadership and social influence This inquiry contributes to communication research and expands the case study method by using the internet as a research tool. This approach challenges traditio nal anthropological notions of field site, moving from localized space to technologically mediated i nteraction, and advances communications literature by offering fresh insight


21 into understanding relationships that develop among participants in online spaces. This research provide s new evidence for how people experience a particular phenomenon, how they gather, share, and act on information related to that phenomenon, and how traditional notions of leadership, trust and credibility are created and maintained in virtual place. This study also offers renewed commentary on what is com monly referred mmunity, Background The Internet and Information Seeking related experiences impacted by amorphous opinion leaders and online group processes are at the cor e of this research, those experiences and processes are set against a backdrop of technology. Clearly, technologically mediated communication is opening up an amazing array of information options for people to access, particularly through the internet and world wide web. But to see this as primarily about the ability to more easily access text or video files is to be somewhat shortsighted. Information access on the internet is a remarkable innovation, but the promise of digital information revolution goes f ar beyond web surfing. The digital information revolution is extending to almost all aspects of daily life, all parts of society, all organizations and most nations (Morris & Ogan, 1996). Without question, many of the ways in which life and society work to day are based on information making the world intelligent with more information is the key to improved quality of life and social progress (Atkinson, 2008). In a world saturated with information, and with the tools to effectively assemble and process it, m odern life is entering a new era where some say information technology is the chief driver of progress and change


22 not only in personal lives and society, but in human civilization (Postman, 1992; M. R. Smith & Marx, 1994). Recent research has revealed ne w understandings in the basic ways people interact, offering new insights into intrapersonal and interpersonal dynamics, group functioning, the development and impact of social networks, organizational behavior, commerce, and global information sharing (Bo ok, 2008; Constant et al., 1996; Eveland, 2003; Kozinets, 2002; Preece & Ghozati, 2001; Rafaeli & Raban, 2005). Study of the internet has reinforced some simple truths and expanded their vitality in illuminating ways: people sometimes relate better as grou ps rather than as individuals; people relate to one another even in the absence of face to face contact; people help each other with & Hancock, 2005). Ultimately the the orizing, speculation, generalization, and hypothesizing about the internet condense to how and why people actually use it. Users have increasingly come to view the internet as a utility rather than a novelty, and report that the internet has enhanced their lives in multiple ways, and has significantly improved communication with friends and family (Rainie & Horrigan, 2005). While this may solidify bonds between and among users, the internet also appears to be a bridging agent for creating and sustaining com munity. Pew Internet and American Life Project (2008) found that 84% of American adult internet users, or approximately 100 million people, belong to groups or communities that h ave an online presence (Table 1 1). Community members said using the internet this way increases opportunity to interact with people outside their social class, racial group or generational cohort.


23 The communal aspects of the internet are key components in this research in that one of its objectives is to understand how community me mbers experience knee arthroplasty (or its potential) and the communication both on and off line surrounding it. relationship is both reflexive and ongoing, with people constan tly reshaping the internet information that can found and shared with relative ease, makes the medium an integral part of modern life (S. Fox & Jones, 2009). It is sa fe to assume that the bond people have with communication technologies at least those who are connected will likely strengthen as connectivity increases through the use of laptops, cellular telephones, and various other personal digital devices, and the co mmunal possibilities of digital media technologies expand. As a communication tool, broadband technology enables people to more easily maintain a greater number of relationships due to its speed and convenience, and the control it offers in managing commu nication (Horrigan, 2008). People use their social networks to seek information and advice, and networks and their structures may provide access to various types of information and resources that offer help, broa dband technologies make these networks more effective (S. Fox, 2006). makes internet access in effect instantaneous, and this feature as a component of broadband technology plays an important supporting role in the maintenance and cultivatio n of social networks (Rainie & Horrigan, 2002). Availability of


24 the technology and the proximity to information that broadband provides changes the way users find, generate, and manipulate content, such as searching for and retrieving health care informati on from the internet, taking an online course, or working from home. According to the Pew internet & American Life Project (2008), people create and manage online content; download, recombine, manipulate, and share content with others. They satisfy a wider range of information searches due to the presence of a high speed connection. purported by many to be profuse. Broadband internet access is claimed by some to be essenti al to stimulating an information society and is professed to be a key driver of both economic and intellectual growth (Benton, 2008; Martin, 2008). The ability to share greater amounts of information at increasing speeds boosts productivity, facilitates in terstate commerce, and helps drive innovation. Broadband has the potential to affect almost every aspect of life in both developed and developing countries by providing new information technology applications, improvements to existing ones, and new service s, investment and job opportunities (Martin, 2008). Broadband can also have considerable impact on everyday life, particularly with regard to telemedicine and eHealth applications, eGovernment education, and rural development (Horrigan, 2008). Although p hysicians rarely make house calls today, traditional health care still relies to a large extent on the physical visit of a patient to a health care provider. But such visits are potentially inconvenient, painful, costly, or sometimes impossible, particular ly for very ill, elderly, or disabled persons, or those living in rural or low income areas where doctors and clinics are scarce. Telehealth the


25 use of telecommunications technology such as broadband in the provision of health care is changing the ways in which medical care is provided (Rintels, 2008). Telehealth technologies now remotely monitor patients, facilitate collaboration between medical professionals, exchange medical data and images, and instantaneously provide efficient emergency service to remo te areas (J. A. Schwartz, 2008). Widespread, low cost dissemination of health information to patients and consumers and reduced travel to and among medical offices, clinics and hospitals are among other benefits of broadband. Broadband technology is also a facilitator of social inclusion, its proponents claim, and thus has considerable impact on a great many people worldwide, including their abilities to form and experience virtual community (Horrigan & Rainie, 2002). As the U.S. and European Union (EU) con tinue to design and implement inclusive broadband internet policies, provisions therein will arguably have direct effects on the creation, maintenance, and likely proliferation of most forms of social media including health based virtual communities. Betwe en 2006 and 2008 alone, activity on social networking sites and in online communities grew 93% (Burns, 2008). Seeking Health Information Online The diffusion of the internet has radically expanded the readily available sources for information of all type s. Roughly 91% of Americans recognize the internet as an ) and a growing number of Americans have reported using the internet and world wide web to find health information in recent years ( S. Fox, 2 005; S. Fox & Jones, 2009; Rains, 2007 ; J. A. Schwartz, 2008 ). Cl ine and Haynes (2001) cataloged some of the advantages of seeking information online: widespread access, anonymity, potential for interactivity,


26 social support, and the ability to tailor info it is indisputable that a plethora of information is available online. H ealth care information is an example of this expansion of important sources, since it includes not only a wide array of documents (ranging from popula r to scientific), but also acts as portal to adv ice from a number of interpersonal sources from laypersons to experts. Those seeking health information have historically been able to consult a variety of sources b ut the internet puts a range of sour ces within reach via a few keystrokes (Case et al., 2004). The physician, the expert researcher and the comp assionate friend are equi digital device. Unlike traditional media books, journals, magazines, ne w spapers, television and radio health informati on on the internet may not be filtered through a gatekeeper, and users are making decisions about informa whether the source is a pharmaceutical company page or a listserv thread (Broadway, 2005). Twenty years ago medical information was likely obtained (if at all) directly from health care professionals, or indirectly through friends or family members who were accessible and perceived to be opinion leaders (Case et al 2004). Now information about diseases, treatments, and prevention is readily available over the internet ; consumers may bypass entirely some of the more traditional sources of health related information. Rains (2007) reports that a growing distrust in health care pr ovision and with entertainment oriented media as sources of health information has increased the related specifics. P revious


27 research (S. Fox, 2005; S. Fox & Jones, 2009; Pew, 2008) has sugg ested that online health seekers were often motivated to search out information relating to actions they may need to take for specific medical issues in their lives. In many cases, online health seekers were action orient ed and highly purposeful due to a p ressing me dical issue for them to address either for thems elves or for another (Table1 1 ). Google has become the de facto second opinion for patients seeking further information p articularly after a diagnosis (S. Fox, 2008). A Google search for a certai n disease or malady will summon a Wikipedia entry and information sites from major hospitals, drug companies and other providers. Add ing o that search displays pages fi lled with sites that connect patients, information seekers, vendors, sufferers and their proxies. Once connected, community website users learn as much or more from each other than from sites offering information alone. This shift is viewed by many as the democratizat ion of health care (J. R. Schwartz, 2008). These expanded capabilities allow people to easily share information, challenging the traditional top down flow between doctors and patients. It is inarguable that health related sites on the internet and world wide web are abundant. According to J. R. Schwartz (2008), these sites can be grouped into five broad, but often overlapping, categories: G ENERAL INTEREST Websites like WebMD (webmd.com), Discovery Health (health.discovery.com) and The New Yor k Times (nytimes.com/health) provide health news, lifestyle advice, and information about disease, as do medical institutions like the Mayo Clinic (mayoclinic.com). M EDICAL RESEARCH These offer access to the published work of scientists, to studies and to portals into continuing research. Examples include PubMed ( ncbi.nlm.nih.gov/pubmed ) from the National Library of Medicine; clinicaltrials.gov


28 which trac ks federally financed studies; PsycI nfo ( apa.org/ psycinfo ), a trove of psychological literature; and the National Center for Complementary and alternative medicine research. P ATIENT SITES Websites for groups and individuals are proliferating, so much so that they are increasingly used by researchers to find patients for studies. These sites include the Association of Cancer Online Resources (acor.org) and e patients (e patients.net), as well as Patients Like Me and Trusera ( trusera.com ), both of which provide Facebook style social connectivity for users, along wi th the ability to share stories in clinical, data laden detail. D ISEASE SPECIFIC These sites focus on a particular condition and are often sponsored by major organizations like the American Heart Association (americanheart.org), the American Cancer Socie ty ( cancer.org ) and the American Diabetes Association ( diabetes.org ). S maller groups however, can put together extensive resources as well, with sites like breastcancer.or g and Diabetes Mine (diabetesmine.com), which calls itsel and BoneSmart. org, the subject of this investigation. W EB TOOLS Such sites designed to help people manage their conditions : S ugarstats.com for diabetes, Destination Rx (drx.com) for comparing drug prices, and YourDiseaseRisk.com, a service of the Washington University School of M edicine that helps patients determine their risk for various problems. One concern the medical establishment expresses about online health seekers is whether they are self diagnosing and/or self medicating based on information they find online and without co nsultation with medical experts (S. Fox, 2006). While it is probable that people do not discuss every book, magazine article, or health related conversation they have wi th their doctor, interest in the typical online health information sess ion persists due to what the medical establishment sees as the largely unfiltered/unedited nature of that information. In addressing this, S. Fox (2006) found t hat 33% of health seekers talked with a doctor or other health professional about the information they found online after their most recent internet search. Later, S. Fox (2008) found that, at least for health, the impact o f an online information search was more likely to be hel pful than harmful. Thir ty one percent of


29 online health information seekers said they or someone they know had been helped by following medical advice or h ealth information found on the internet. Just 3% said they or someone they knew had been seriously har med by follo wing advice or information they found online. This sheds some light on the subject of principal sources of health information. While research indicates that medical professionals are the dominant source for urgent health questions (S. Fox, 200 5, 2006; S. Fox & Jones, 2009; Pew 2009), evidence continues to mount demonstrating that consumers are taking advantage of the convenience of the internet as an initia l source of disease information, and reliance on information from other sources at least at the outset of an information search appears to be falling (Case et al., 2004; S. Fox, 2010; Hesse, Moser, & Rutten, 2010). In theory, this should have implications for the concept of opinion leadership as it pertains to health care and the internet. It is not just convenience that draws health information seekers to the internet but th e positive experiences they have with online research and interaction. T he kinds of health information sought and found online are different from what people can glean fr om most traditional sources and this translates into much more than doctor patient communication (S. Fox, 2010). The internet is a medium that health seekers can tap into wherever and whenever they need it, connecting with whatever source they believe wil l help them at that moment. Today, t ha t source is not always a doctor, or even a health professional.


30 Joint Replacement Around the world, millions of people have conditions that affect bones and joints. Because people are living longer, medical conditio ns that affect these parts of the human anatomy, like arthritis and osteoporosis and which arise from injury, genetics and/or aging, have a great impact on co sts of care and quality of life (Losina et al., 2009). The knee is the largest joint in the human body as well as one of the most vu lnerable. Normal knee function is required to p erform most everyday activities; the knee joint must bear the weight of the upper body as well as the stresses and shocks carried upward through the feet when a person walks or runs ( R ighthealth.com 2010 ). Knee replacement surgery, or knee arthroplasty the most com mon joint replacement procedure today, is a surgical procedure most often performed to relieve the pain and disabili ty from degenerative arthritis mainly osteoar thritis but it resolves other arthritides as well ( Kurtz, Ong, Lau, Mowat, & Halpern, 2007). Knee arthroplasty has two primary purposes: pain relief and improved functioning of the knee joint. The U.S. Department of Health and Human Services (2009) report s approximately 547 ,000 knee replacement operations are performed each year in the U.S Althoug h about 70% are implemented in people over the age of 65 (j oint diseases account for more than half of all chronic c onditions in the elderly), a grow ing number a re being done in younger patients. Currently over 11,000 knee replacement procedures are performed annually on U.S. patients 15 44 years of age (DeFrances, Lucas, Buie, & Golosinskiy, 2008). R easo ns for this trend include improved su rgical techniques and i nnovations in the design and construction of knee prostheses since the first knee re placement was performed in 1968 (U.S. Department of Health and Human Services, 2009 ). Moreover, people's attitudes toward aging and their e xpectations of an active life are changing.


31 Fewer are willing to endure years of discomfort or resign themselves to restricted levels of activity (Grindrod et al., 2010). With one thir d of Americans obese, coupled with an aging population the prevalence of osteoarthritis is expected to increase, contributing to a growing demand for joi nt replacement procedures ( U.S. Department of Health and Human Services 2009, p. 76). B y 2030 the demand for total hip replacements i s estimated to increase by 175% and the demand for total knee replaceme nts is projected to grow 600% (Kurtz et al., 2007 ). i nterest in knee arthroplasty goes beyond the mere fascination of replacing human body parts with man made simulations that actually work and enhance q uality of life for recipients. Th e experience of how one journey s from initial recognition of the condition through diagnosis, reference group interacti on, information search, introspection, family deliberation, and eventual decision, is worthy of study alone. Yet when forks in the cognit ive road on that journey are undertaken through computer mediated interaction, much can be learned about patterns of human variants of these processes might operate for not j ust the elderly, but for example, cheerleaders, firemen, snowboarders, grandparents, athletes, single working mothers, perception of their condition? T he internet has inar guably become an important source for health related information. At this writing, a Google In addition to seeking health


32 related information or advi ce on the internet pages of professional health care organizations, or seeking professional advice in the form of online consults, people increasingly make use of online discussion forums organized around health related topics (S. Fox & Jones, 2009). In th ese forums also referred to as self help groups social support groups or in some cases virtual communities (Rheingold, 1993), people can exchange information, ask for help, discuss problems, and tell stories to others who, at least to some extent, have experience d or have an understanding of what others here are going through. line network largely consists of family, friends and colleagues, online forums m ay facilitate contacts more diverse in their social, cultural, or geographical bac kground s (Wright & Bell, 2003). Thus, a greater diversity of relations hips (from close friends to total strangers) may provide people with relevant and potentially more diverse information since these sources would be less accessible (or inaccessible) offl ine (Ric e & Katz, 2001; Wellman, 1997). This information can be use ful le with more control and reduce uncertainty, which in turn can help people in copin g with their particular situation. Having first hand e xperience with the topic at hand, many users in these forums are increases their potential for influence (Tanis, 2008). Newer members in these communities glean information about how to behave or cope wi th the situation they face modeling the experts by way of social comparison (Davison, Pennebaker, & Dickerson, 2000). This topic was chosen bec ause: 1) 83% of American adult internet users (61 % of all U.S. adults) have searched for health information online (Pew Internet & A merican


33 Li fe Project 2008), which makes it important; 2) knee arthroplasty is a major investment which calls for informed research; and 3) knee replacement is not l ife threatening, which c r eates conditions whereby members may not feel as pressured to remain in the community, as if they would in, say, a virtual community formed as a cancer support group. This may allow for a more wide ranging and open exchange of information, including off t opic discussions. BoneSmart.org This community was chosen because it had a diverse population of users that has remained relatively stable over its eight year existence. Given that a wide variety of information sources exist within the boundaries of this community to help members discern information quality and facilitate interact ion with other participants in meaningful ways, individuals here seek real solutions for real problems and are concerned with presenting themselves in ways which convey the identi ty that instigated his or her membership, rather than inventing new ones a research concern in many virtu al communities (Turkle, 1995). The BoneSmart Community offers an opportunity to research how members here gather and make meaning of not only informat ion they share on wide ranging aspects of knee injury and potential replacement, but also on anything members feel may be important enough to share. BoneSmart.org is an orthopedic implant and joint replac ement consumer awareness website established in 200 2. Th e primary goal of the BoneSmart Community Forum is to foster communication between persons who may need joint replacement surgery and those who have had one or more The site encourage s ex new patients, and provide a


34 nurturing environment of su pport and knowledge through the community forum (BoneSmart.org, 2010). While the site also supports a Hip Replacement Forum, BoneSmart members, offers comprehensive infor mation regarding knee pain, injury, replacement and surgery recovery, and provides a network for its members to discuss a variety of on topic issues. This site also provides webcasts, links to reference and video libraries, and step by step preparation gui des for any and all forms of knee surgery. Only members can post information (registration is required). Research in Medicine, Inc.) is a 501(c)(3) non profit public benefit entity A 12 member Founders Group (which serve s provided initial funding for FARM and s ustainable funding for the site come s from annual network fees, corporate sponsorships, ind ividual contributions and fundraising a ctivities. BoneSm art is the public awareness arm of FARM Orthopedics, described as national campaign to make consumers aware of new advances in orthopedic FARM Orthopedics enters strategic alliance partnership arrangements to provide products and ser vices to its network ; BoneSmart supporters include manufacturers of medical devices, research and development institutions, physicians, professional organizations, hospitals, clinics and private individuals. FARM Orthopedics also provides developme nt funds and marketing resources to various entities that support research, professional development, product innovation and consumer awareness programs. The Swiss Colony MassMutua l and Canyon National Ba nk contribute a portion of consumer


35 purchases to the Foundation in the form of donations. Sponsor advertising is apparent In addition to questions of opinion leadership and influence, other important questions to ask in this investigation might be: How do community members situate communication technologies in their daily lives? How do these members evaluate the information they glean from virtual community? How are the tools of new media ch anging communicative practices? How might new technologies alter novice expert relations in this case as between community members and the medical establishment? This research helps to understand the ways in which people incorporate new communication techn ologies into existing communicative routines, and how these technologies influence new and emerging cultural practices. Further Chapters Chapter 2 describes scholarly research relating to the theoretical underpinnings of this study, exploring the strength of social ties, virtual community, opinion leadership the two step flow of communication, trust, credibility and homophily. The chapter also poses the research questions this study sought to answer Chapter 3 details the methods used which include Soci al Network Analysis, an online questionnaire, e mail interviews, and explains procedures and method s of data an alysis. Chapter 4 summarizes data resulting from these mixed methods procedures and resolves the research questions, while Chapter 5 draws conclu sions and offers interpretations based on findings. Chapter 5 also di and proposes areas for future research.


36 Trade association or professional group _________________ _____ 50% People who share a hobby or inte rest _________________ _____ 50% Fan group for a particular sports team ____________________ 31% Fan group of a T V show or entertainer ___________________ 29% Local co mmunity group or association ____________ _______ 29% People who share the same lifestyle __________________ 28% Support group for medical condition or personal problem ___________________ 28% Peopl e who share the same beliefs _________________ 24% A political group ________________ 22% A religious organization _______________ 21% Sports team or league in which member participa tes ______________ 20% Ethnic or cultural group ____________ 15% Labor union _______ 6% Figure 1 1. Percentage of American adult internet users who say they belong to this type online g roup. Note: Data in figure 1 are from Pew Internet and American Life P roject 20 08, latest trends, updated July, 2008). Table 1 1. Percentage of internet users who look online fo r specific health topics % Health topics 66 a specific disease or medical problem 55 certain medical treatment or procedure 52 exercise or fitness 47 doctors or other health professionals 45 prescription or over the counter drugs 38 hospita ls or other medical facilities 37 health insurance, including private and Medicare or Medicaid 35 alternative treatments or medicines 33 how to lose weight or how to control weight 28 depression, anxiety, stress, or mental health issues 26 any other h ealth issue 20 experimental treatments or medicines 12 how to stay health on an overseas trip Note: 83% of internet users (61% of American adults) have looked online for information about at least one of these topics. Source: Pew Internet &Ame rican Life Project, The Social Life of Health Information 2009, p. 11.


37 CHAPTER 2 REVIEW OF THE LITERA TURE This chapter provides an overview of literature pertinent to an investigation of how people seeking information for a specific health through technologically mediated interaction and how first hand experience using online online. Also relevant is literature pertaining to how patterns of influence operate in virtual communities and how the social processes that create and maintain these communities function. Strength of Weak Ties As digital media technologies continue to transform mass communications, there is mounting e vidence that the communications revolution is also changing the ways in which humans interact (Gillmor, 2006; Jenkins, 2006; Luders, 2008). The communication technologies that make use of the internet and world wide web offer both possibilities and constra ints for communicative practices and social interaction, and provide a context for emerging forms of communication in the nexus of psychology, sociology, technology and science. Rooted in 1982 ) strength of weak ties theory has implications for this research in that its premise relative strangers could offer an advantage over friends and colleagues in obtaining useful information takes on added significance in a new media environment. Granovetter argues that strong tie relati onships occur among people who are similar in many respects; similar people are more likely to know the same things and less likely to know dissimilar things. When information is unavailable through strong ties, people may obtain it through weak ties:


38 rela tionships characterized by infrequent contact, lack of emotional closeness, and no history of reciprocity. Weak ties serve as information bridges across cliques of strong ties and can offer people access to resources not found in their strong tie relations hips. 1361). In a labor market study of professional, technical, and managerial jo b changers in a Boston suburb, Granovetter (1973) found that of those finding a job through interpersonal contacts, 16.7% reported that they saw their contact often at the time they sought employment, 55.6% said occasionally, and 27.8% rarely (p. 1372). Th is would indicate the successful contact was someone only marginally included in the job such as an old college friend or a former work mate or employer, someone wit h whom only sporadic contact had been maintained. From an individual's point of view, then, weak ties are an important resource for social mobility. Seen from a macro perspective however, weak ties would appear to play a role in affecting social cohesion. When someone changes jobs, she is not only moving from one network of ties to another, but also establishing links between them. These links are often of the same kind which facilitated her own movement (p. 1373). Subsequent research has lent support to th e strength of weak ties theory. In a study on the diffusive nature of suicidal thoughts in the friendship networks of adolescents, Baller and Richardson (2009) found that "at risk" respondents those with certain behaviors, statuses, and experiences that cr eate psychological predispositions


39 to suicide are more likely than those deemed not at risk to seriously think about committing suicide when a friend of a friend attempts suicide. Looking at peer effects, network structure, and juvenile delinquency, Patacc hini and Zenou (2008) found that increasing the number of weak ties in the network induces more transitions from non crime to crime, resulting in an increase in the crime rate. These researchers reported that as the number of weak ties increases, as measur ed through friends of friends (including both delinquents and non delinquents) the crime rate rose. Relating the strength of weak ties theory to computer mediated communication, Constant et al. (1996) used a computer network in an organizational setting t o draw on weak ties linking people across distance, time, country, hierarchical level, and organizational subunit. Information seekers sought out technical information or referrals to solve technical problems; the researchers found that information provide rs gave useful advice and solved the problems of information seekers despite both a lack of personal connection to the seekers and organizational motivation. While the diversity of ties as well as the resources of information providers contributed to wheth er or not and to problem solving. The researchers concluded these ties were useful to the degree they put people in touch with those offering superior resources, but they were not useful nor did they have a greater likelihood of solving the information seeker's problem when they were simply greater in number. Further investigating the strength of weak ties and the internet, Kavanaugh, Reese, Carroll, and Rosson (2005) fou nd that the internet helps to increase the number of weak ties across social groups. Focusing on social cohesion in a mid sized


40 community with high internet penetration, surveys and interviews helped these researchers determine that people with a greater n umber of weak ties across groups have higher levels of community involvement, civic interest, and collective efficacy than people with fewer. Moreover, heavy internet users with more weak ties had higher social engagement, used the internet more for social purposes, and attended more local meetings and events since going online than did heavy internet users with fewer weak ties. Kavanaugh and colleagues concluded that in this context, by enhancing the flow of information across diverse members of the commun ity, the internet raises awareness, educates interested citizens about complex local issues or concerns, and sometimes mobilizes individuals and groups for collective action (p. 129). They also contend it is the usual suspects, community leaders and active participants, who are using the internet to distribute information of civic interest, a finding that has implications for one concept under study in this research opinion leadership. Arguments have been proposed for why weak ties are useful. One is that weak ties comprise more numerous potential helpers than do strong ties; if weak ties are more numerous than strong ties, then calling on weak ties increases the probability of acquiring useful information (Friedkin, 1982). Yet perhaps it is not the number of people giving advice that makes weak tie contacts useful, but the range or diversity of those ties. Burt (1983) points out that if all of a person's weak tie contacts are themselves members of the same strong tie group, then the expertise offered by tho se ties will be redundant. Burt suggests it is the extent to which weak ties tap diverse groups that makes them useful. Group diversity could also increase alternative solutions (people offer different answers) or provide pieces to a multi part solution (p eople offer partial


41 available help across a variety of topics, greater advantage may be realized by es. Prior to the internet, weak tie networks for most people consisted of neighbors, service providers, acquaintances and other lower intensity relationships that a person might turn to for information or support when closer ties (e.g. friends and family m embers) were unavailable (Granovetter, 1973). The internet has greatly expanded the number of contacts that could potentially become weak tie support networks for people with health concerns. According to Wellman (1997) weak tie networks in computer mediat ed environments tend to be more heterogeneous than strong tie networks, and weak ties are usually better connected to other, more diverse social circles, and hence more apt to be sources of new information. Closer ties tend to be more homogenous because pe ople tend to form relationships with others based upon proximity, demographic background and attitudinal disposition (J. C. McCroskey, Richmond, & Daly, 1975). lack of social prese nce online fosters relationships with people who have more diverse presence is limited by reduced non verbal cues (physical and social) in the computer mediated environme nt, which allows relationships to develop on the basis of shared interests rather than be cut short at the onset by perceived differences arising out of face to face contact. arch in that the internet now helps people develop and maintain weak ties within and across


42 large and diversified networks. Rather than relying on a single community for information, advice, and resources, people have success when they actively seek out a variety of appropriate alternatives in different situations. The macro view reveals how the personal experiences of an individual are bound up with larger scale aspects of social structure; how the relative strength of interpersonal ties in and across grou ps work to form macro patterns in society, patterns that reveal how people search for information, how they use the internet, how they solve their health problems, and how they create community in virtual space. Virtual Community Rising numbers of people are spending considerable time in virtual space, space that does not occupy a geographical locale, but is wholly mediated through technology by way of a material object. Intimacies with machines are creating a world in which it makes sense to consider a ne social existence via technology, the new spaces they inhabit suggest a revised placement of the human subject (Wellman, 1997; Wellman & Gulia, 1999). Virtual community encompasses an assortment of n otions, including how people perceive online space historical, geographical, practical and sp atial understanding of community shape s their perception of virtual community), which kinds of gatherings constitute community in cyberspace, and the nature of shared commonalities. Factors that affect whether and how one identifies oneself as a member of a virtual community include trust and commitment, intensity of the social relationships that develop out of online interaction, and whether a n individual relates to the social environment as seek social support in online community and while early internet research revealed instances of identity manipulation, recent studies show


43 that people are now more interested in frank online representation, particularly in health related environments. Th ese notions of community, commonality, identity, space and place, social support, and online health community will therefore be explored in this section. Community Markham (1998) sugges ted that internet users frame their experiences of computer mediated communication along a continuum of connection of self, and generated three themes along this continuum to help understand online experiences. First, computer mediated communication is a tool that facilitates communication itself (a decidedly materialistic observation); second, cyberspace is a place to go to be with others; and third, computer mediated communication is a way of being that is inseparably woven into lived experience. Th ese notions have implications for where individuals see themselves performing online actions. According to R. Park (1936), the essential characteristics of a community are: 1) a population territorially organized; 2) said population more or less comple tely rooted in the soil it occupies; 3) a population in which its individual units live in a relationship of community has been continually redefined and remains extraordinarily slipper y. A more recent geographic area, common ties through identification by residents with one another and with that area, and substantial social interaction among the re When considering virtual communities, conceived and practiced in amorphous community rooted in the soil may seem outdated in a modern socie ty, but traditional


44 ideas of community that identify with place remain strong (Grannis, 1998; Molotch, Freudenburg, & Paulsen, 2000). Identification with geographical place would seem to exclude virtual community from these definitions. The term communi ty, when used to describe virtual communities, is contentious. A through computer supported social networks within a social structure in cyberspace (Driskell & Lyon, 2002, p. 375). If the medium is a computer network as this definition implies, the community can be called an online community. Virtual communities are usually (but not always) dispersed geographically and are therefore not communities under earlier definitions are print based virtual communities; although ephemeral, a film crew may be considered as such, as it gathers in a ph ysical location to perform a specific task, displays the character of a place based community for a length of time, and once the task at hand is complete it disperses and gathers elsewhere possibly on the other side of the planet. Some online communities are linked geographically and are known as community websites. However, if one considers communities to simply possess boundaries of some sort between members and non members, then a virtual community is certainly a community. The idea of neatly bounded co mmunities is also being critiqued, since communities are just as fluid as they are static, with members joining and leaving, even being part of different communities simultaneously (Wellman & Gulia, 1999). The term virtual community stems from the title of a 1993 book by Howard Rheingold, in which the author chronicles his adventures in a number of computer mediated communication


45 and social groups. In its pages Rheingold offered what he considered the potential benefits of belonging to such groups, describi ng what these benefits may mean both for him personally and for society at large. Today the terms virtual community and online community are used for a variety of social and professional groups interacting via the internet. It does not necessarily hold th at there are strong bonds among members, although Rheingold (1993) mentions that virtual communities form "when people carry on public discussions long enough, with sufficient human feeling, to form webs of personal relationships" (p. 51). An e mail distri bution list may have hundreds of members and the communication which takes place may merely be informational (questions asked, answers posted), but members may remain relative strangers and membership turnover rate could be high. This is in line with a lib eral use of the term community. Commonality The internet certainly provides opportunity for associations based on all genres of shared interests. Yet some researchers (i.e., Etzioni & Etzioni, 2001; Galston, 2000) contend that newsgroups, mailing lists, f an clubs or chat rooms do not constitute community because a true community must have more in common than a narrowly defined topic. Common ties require commonalities; groups of individuals should have a bond, a measure of commitment, a set of shared values a culture, a history, a shared identity (Wellman & Gulia, 1999). Cyberspace communities may foster specific communities of interest rather than communities of multi interests, the type which would generate and support personal connections and broad famil iarity among members. Although associations with others through the internet are key to the concept of


46 (Ridings & Gefen, 2004). Some chat rooms and discussion groups for instance, are but places for people to congregate without a sense of permanence or consistency among the site patrons. And chat rooms largely cater to singles looking to meet other singles, each room containing a different mix of people each day, non e necessarily returning on a regular basis. Such chat rooms and groups, because they lack a regular basis of participation by their patrons, do not qualify as virtual community (Q. Jones, 1997). Ridings et al. (2002) offer a definition of virtual communit y that will be utilized in this practices that [sic] communicate regularly and for some duration in an organized way 273). Kozinets (1999) defined virtual communities as affiliative groups whose online interactions are based upon shared enthusiasm for and knowledge of a specific consumption activity or related group of activities (p. 254). Meta analyses of computer med iated communication indicate that group members generally progress from initial social information gathering to increasingly affiliative social activities (Walther, 1995). At first, potential members may merely browse information sources, lurking (inconspi cuously reading, but not posting) to learn about an activity or interest. An subject. As th is user becomes more sophisticated in her internet use and more knowledgeable about the topic, she will begin to visit sites that offer more specific information, and eventually make online contact with other consumers of knee replacement/surgery informati


47 query individuals, or pose questions to the entire community. Her level of interest in accumulating this information and whether she is gratified in obtaining it from this source will dictate whether sh e becomes a frequent or occasional participant in group discussions. In general, a virtual community member will progress from being a visitor to an insider as she gains online experience and discovers groups whose activities satisfy her needs (Kozinets, 1 999). Identity How one identifies him or herself as a member of a virtual community depends largely on two interdependent factors. First is the relationship the individual has with the activity or subject in question. The more central the activity to the p psychological self self image, the more likely the individual will pursue and value membership in a community (virtual or face to 2001). Second is the intensity of the social relationships one develops with other members of the community. The quality of social interaction is highly dependent on trust and intimacy; virtual communities may lack this psychological closeness as many fe atures of virtual community do not promote interpersonal knowledge, trust or commitment (Galston, 2000). While members in a traditional, local, place based community will have a degree of social responsibility, the virtual community ostensibly requires no communicate online in ways they probably would not in either face to face or physically proximate circumstances (i.e., telephone or videoconferencing) (Turkle, 1995). Self representa tion is and has always been an issue in the study of the computer mediated communication. While the influence of the internet on self representation has


48 been under investigation for almost two decades, most early studies focused on anonymous environments s uch as MUDs (multi user dungeons), chat rooms, and user groups (Rheingold, 1994; Turkle, 1995). These studies found that individuals tended to act at being someone else or act out their underlying negative impulses in the online world and pointed to the op portunities a nascent cyberspace offered for becoming emerged through potenti al for anonymity in a context of text based interaction that circumvents basic identity cues (such as race, gender, age) represented by the body. Both these approaches incorporate properties of technology and the absence of a corporeal body in the interact ion into explanations of how and why people play with identity. As detailed empirical, ethnographic and biographical studies began to emerge, different conceptualizations of internet identity now appear on the theoretical landscape. More recently, research ers have shifted their attention to self presentations in less anonymous online environments such as internet dating sites (R. Jones & Ortlieb, 2006) and health information users groups (T. Lewis, 2006; Tanis, 2008). Results here suggest that people presen ted themselves and acted differently in such environments than they did in other online settings. These and similar findings (Kennedy, 2006; Preece & Ghozati, 2001; Wright, 2000; Wright & Bell, 2003) indicated that the online world was indeed not monolithi c, and online identity construction varied according to the nature of the setting. With the emergence and widespread use of the world wide web as a display technology, presentation of the self could now include more defining elements


49 of identity such as ph otographs, audio and video files, and links to others sites for verification which suggests that some displays of online identity point toward a more rigid and static use of the internet and world wide web. Carter (2004) concluded that despite claims as to multiple and fluid identities in virtual space, residents of her The internet and world wide web user population is far different than what was revealed in the early landmark studie s (i.e., Lea & Spears, 1992; Rheingold, 1994; Turkle, 1995). A more equitable gender split and increased racial diversity are apparent (Pew, 2008) and the demographic shift in the user population has been accompanied by drastically reduced interest in onli ne role playing as a percentage of the total internet use (Kennedy, 2006). Today users go online for many different reasons and online identities are likely to be extensions of offline identities because for most people, internet use enhances, extends, and supplements what they do offline (Rainie & Horrigan, 2005). This suggests that rather than inventing new identities, the majority of those who use the internet and world wide web hold on to those identity components that are most important to them, while searching the web for information or answers to their problems. Space and Place So for virtual community the question remains: Does geographic place still matter? Confining community solely to spatial parameters neglects the relational elements common ties and social interactions necessary for community, but some observers argue that just because people are neighbors does not necessarily mean they are friends (Fischer, 1982). Local place is not always sufficient to produce community, but is it necessary?


50 A debate developed within earlier research on internet communication about whether online, virtual, or otherwise computer mediated communities were real or imagined (Bourdieu & Colemen, 1991; Markham, 1998). At issue was whether these sorts of communities w ere too ephemeral to explore as communities per se, or whether the nature of the communication medium made them radically different from what had heretofore been thought of as communities. While face to face contact may be of technology enabled communication channels has diminished this traditional method of social interaction (Dwyer, 2007, p. 2). Increasing use of cellular telephones, text messaging, instant messenger, e mail and social networking sites indicate that digit ally mediated communication is a n important factor in the management of interpersonal relationships, and it is out of these relationships that virtual communities congeal into place (Carter, 2004). Thus the distinction between real and imagined or virtua l community may be less concrete than it seems, regardless of the ways in which community members interact. What remains elusive is how people generate a sense of place that is different from the spatial logic of physical space. Familiarity with any virtu al community produces the notion that it is a specialized place within a landscape that contains any number of recognizable social spaces. As such, it is a place that allows members to invest in social interaction in ways similar to how they would in other social spaces with which they are familiar. If, as Lefebvre (1991) argues, space is produced through social action rather than being a priori, then the place that is a virtual community (or social network) becomes a social product. Hence, when users dis in space where


51 collective and individual practice creates place. As such, it is not a technological construct but a cultur al construct, mediated through experience rather than through technology (Carter, 2004, p. 113). Tilley (1994) brought space further into phenomenological inquiry by contending that space has no fundamental essence in itself, only relational significance c reated through relations between people and specific spaces. These social spaces, rather than homogenous and static, consist of varying densities of human experience, connection and participation. They are contextually constituted, and provide particular s ettings for involvement and the creation of shared meaning (Tilley, 1994, p. 11). Hence space is transformed into place. This transformation is key to understanding how space virtual space for this study becomes place for subjects of this research. Pla ces are created out of space by groups or individuals through shared symbols, common experiences and common meanings (R. Jones & Ortlieb, 2006). Place is both internal and external to a human subject, at once both a center for personally embedded meanings and physical action. Basso (1996) contends that place is primary to the creation and maintenance of personal and group identities and through the act of naming, places become invested with meaning and significance for individuals and groups. Naming is a co mponent of familiarity with a place people routinely draw on their stocks of knowledge in the places in which they act, to give meaning to their lives and significance to the place. The act of naming transforms the physical into something that is historica lly and socially experienced. The bestowing of names helps to create place out of space (Basso, 1996).


52 Places, however, possess the capacity not only to trigger acts of self reflection, but also lead to thoughts of other entities places, people, times, ob jects whole networks of associations generated by the place itself (Tilley, 1994). As place stirs the thoughts and feelings of those in them, these same thoughts and feelings animate these inhabited places through the attention placed on them (Basso, 1996 ). This process illustrates that familiar places are experienced as meaningful, their significance and value residing in the form and arrangement of their observable characteristics. People continually perform acts that reproduce and express their sense of place, acts which are inextricably tied to their own understandings of who and what they are (Basso, 1996, p. 57). It has been argued that the world is not external to human action waiting for beings to enter it, it is through human action and interactio n that beings bring about the world they are then in; they create so that they may be in the world (Heidegger, 1977; Richardson, 2007). How people adapt and react to distinctly different circumstances and surroundings depends on the interaction between t hem and the material aspects of each particular set of circumstances. When people enter a physical space, they begin constructing situations defined by their interpretive responses to the material aspects of the setting. When people are engaged, when they are incorporating the material setting into the situation, they are constructing a reality that will be different from other realities they construct when confronted with a different material setting (Richardson, 2007, p. 88). This is surely the case when people sit at their computers, log on to the internet, and interact with friends, strangers, and entities in places they have created out of ungrounded cyberspace,


53 in th e The proposition that nonspatial (in the traditional sense) virtual communities can replace local place and shared community space is questionable. It appears to Driskell and Lyon (2002) that internet relationships can complement aspects of community found in local place and shared space, but they are poor replacements for social relationships among individuals based on shared identity, common values, psychological connections, and kinshi p. Communities still exist most readily, most naturally and most often when people identify with place the neighborhood, the church, the school or the workplace and the personal face to face interactions that occur within boundaries of a geographical area (Driskell & Lyon, 2002, p. 388). Others however, (Carter, 2004; Heidegger, 1977; R. Jones & Ortlieb, 2006) contend that people need a place in which to be before they can act and interaction at any level in a specialized place establishes the connections t hat prime relationships, which then initiate the communal ties that transform into community. As virtual community has no universal definition, the argument might just be the degree to which relationships become personal. Functioning as information neigh borhoods, virtual communities offer members congenial environments, places where information in which they are interested is likely to be found even if they do not have explicit queries. Because virtual communities function within a general context of shar ed interest, participants tend to be aware of information that would be of potential interest to others, and thus can share that information without necessarily going through the formalities of querying an information


54 retrieval system. Additionally, becaus e virtual communities are situated within the broader information environment of the internet as a whole, information related to the interests of participants can be either embedded within the community by posting or offered to the community through hyperl inks (Burnett, 2000). And while these environments may lack the face to face interactions of more traditional communities, they provide considerable opportunity for socializing, for giving and receiving support, and for other types of interaction that supp lements the exchange of information. Social Support To give and receive support is an important aspect of social interaction, and since the internet and world wide web have become so integrated into modern day life, it is no surprise that social support i s increasingly exchanged online (Preece, 2000; Tanis, 2008). While features of digital communications media affect how people interact online how they compose and receive messages, form and maintain relationships, organize and structure groups online commu nities formed to help others with specific issues or problems offer members some unique characteristics compared with face to face groups in dealing with the topic at hand. The anonymity and text based nature of community exchanges allow for physical place and time independent interactions that increase the possibilities for members to expand these supportive connections (Wilson & Peterson, 2002). While there are any number of reasons why people engage the world wide web and the internet, one certainly is for social support (Eichhorn, 2008; Preece, 2000). Individuals can associate with others who may share some similarities or who may have an understanding of their specific situation, anywhere in the world at almost any time (Book, 2008; S. Fox & Jones, 20 09; Preece & Ghozati, 2001). Online communities can


55 that may precede, supplement, or in some cases re place, that of health care professionals (McKinlay & Marceau, 2008). In online communities such as support groups, professional groups, and activity based and topical groups, people interact with various others: those who find themselves in comparable cir cumstances, share a mental or physical condition, have suffered a similar traumatic experience, or at least share a topic of interest. This perceived similarity combined with relatively easy access to large numbers of others can provide one with a communal sense of universality that is not likely to be found offline (Preece & Ghozati, 2001; Wright & Bell, 2003), leading to a notion of group membership and an emergent social identity. Preece and Ghozati also found that in online communities where social supp ort was the main focus, suspicion was virtually non existent, interactions displayed few emotional negative remarks, high levels of emphatic communication were observed, and representations of self went overwhelmingly unchallenged. Online Health Community A virtual community in health care refers to a group of people (and the social structure they collectively create) founded on telecommunications with the purposes of collectively conducting activities related to health care and education. Such activities can include actual delivery of health care services, staff or patient education, a platform for providing support, discussing health and treatment related issues and problems, sharing documents, consulting with experts and sustaining relationships beyond face to face events (Demiris, 2006, p. 179).


56 The BoneSmart knee community, subject of this research, represents a particular type of non anonymous, or onymous, online support forum; a virtual community oriented specifically to foster communication among persons looking for information about joint replacement surgery or who have had one or more joints replaced. This forum also encourages experienced users to help mentor potential and new patients, and aim to provide a nurturing environment of support and knowledge. The critical nature of health information and the design of these type entities accessible, easy to navigate, welcoming websites coupled with voluntary membership in common interest groups that expect and rely on peer education and support for h ealth issues comprise a special orientation and affect the ways in which users present themselves in these communities. Opinion Leadership The importance of personal influence that close social ties play a part in the flow of information from media to mas s became apparent to social scientists after several mid 20th century studies revealed that people received a great deal of their information from other people. The concept of opinion leadership emerged from this acknowledgment, along with recognition of s ome of the variables that made it a viable social cognitive theory, social power, and, the two step flow hypothesis, as well as the In the 1940s and 1950s a small group of researchers framed a theory of public opinion formation that sought to reconcile the role of media influence with the growing realization that in a variety of decision making scenarios, ranging from political to personal, individuals may be influenced more by exposure to each other than to mass


57 media sources (Katz & Lazarsfeld, 1955; Lazarsfeld et al., 1948; Merton, 1949). intermediaries between the mass media and the majority of society. Because opinion leaders to less media active segme nts of the population, Lazarsfeld et al. (1948) called their hypothesis the tw o step flow of communication, in contrast to what at the individuals as defenseless objects of media influence (Baran & Davis, 2000). Studies of personal influence and the two step flow contributed significantly to to be recognized as an important mediating variable in the social flow of information and influence from media to mass. Finding s from this and similar studies (e.g., Hovland, 1951; Klapper, 1960; Lewin & Grabbe, 1948) added to a growing belief among communication scholars of the time that the hypodermic needle theory messages propelled from an all powerful direct effects media cou ld no longer be supported by newer data. In the decades after the introduction of the two step flow, the concept of opinion leaders became central in the literatures of the diffusion of innovations (Rogers, 1995; Valente, 1995), communications research ( J. O. Brown, Broderick, & Lee, 2007; Davis, 2006; Weimann, 1991, 1994), marketing (Chan & Misra 1990; Kozinets, 1999, 2002; Myers & Robertson, 1972), and now health ( D. K. Kim, 2008; Maloney Krichmar & Preece, 2005; Valente & Pumpuang, 2007). By the late 1 960s, the hypothesis had been hailed as one of most important formulations in the behavioral sciences (Arndt, 1967),


58 and by the late 1970s the two sociology (Gitlin, 1978, p. 207). Katz and Lazarsfeld this definition remains in use, more or less unchanged (Shah & Scheufele, 2006). It is important to note th at leadership comes in many forms. Most noticeable are official leaders of formal groups heads of corporations, labor unions, government entities, universities and the like. Informal leadership, however, is less visible. It occurs when people need informat ion, advice, or interpretations and turn to others with whom they special expertise or wisdom regarding the topic about which a decision must be made (Lowery & De Fleur, 199 5). Their influence is direct and derives from their informal Numerous studies, including that of Katz and Lazarsfeld (1955), suggest that opinion leaders and followers alike are exp osed to a mix of interpersonal and media influence (Watts & Dodds, 2007; Weiman n et al., 2007) and that differences in influence are more appropriately described on a continuum rather than dichotomously (Lin, 1973). J. J. Brown and Reingen (1987) found tha t even in a relatively small population, 90% of recommendation chains extended over more than one step and 38% involved at least four individuals, which illustrates the notion that people have multiple partly distinct and partly overlapping groups of acqua intances (Cooley, 1902; Granovetter, 1973; Watts, Dodds, & Newman, 2002).


59 The Primary Group sociologists (e.g., Cooley, 1902, 1909; Hovland, 1951; Lewin & Grabbe, 1948) in the face to face relationships (family, friends, neighbors, acquaintances) that are such an important influence on human behavior. If it were possible for the proverbial (and overworked) man from Mars to take a fresh look at the p eople of Earth, he might be impressed at how much time they spend doing things together in groups. He would note how they cluster in small groups under the same roof satisfying biological needs, rearing children and mutually caring for each other, how they enjoy themselves engaging in recreational sports of various kinds, how they spend time in little groups talking, planning, and conferencing, and how the education and socialization of children tends to occur in larger groups like schools, churches, or oth er social institutions. He might also be saddened to find groups of men engaged in warfare. In short, if he wanted to understand what might be going on here, he would have to carefully examine groups. Understanding how individuals relate to groups, how gro ups relate to larger society, under what conditions groups (and networks) form and dissolve, and what makes them grow and effectively function, are keys to understanding human behavior (Cartwright & Zander, 1968). Primary group research led to the develop ment of theories that rested on the assumption that small groups have a profound influence on nearly every aspect of social life, including how people directly and indirectly attend to messages in the flow of information from media to mass (Cooley, 1909; L ewin, 1948). As advances in technology broadened theoretical ground in the study of human communications, middle range theories from other disciplines impacted communication research. Social


60 cognitive theory (Bandura, 1986), based on observational or vicar ious learning, contends that an individual can gain knowledge by watching others or by processing information given to them by others (or by media). By modeling behavior or performing actions based on shared meaning, people transcend the bounds of their im mediate environment. Not only is modeling an important mechanism for diffusing ideas, values and styles of conduct within societies, it also influences transcultural change (Bandura, 1986, p. xii). Social cognitive theory plays a part in this research thr ough modeling that occurs in the BoneSmart community. Opinion leaders in this community, according to his or her experiences, pass on information to other members, who may then base potentially life changing decisions on that information. When the informa tion advocates an activity that requires an investment of time and/or resources, and failure can be costly (very costly for this condition, and not in just monetary terms) people seek verification from multiple sources before they act. These activities mig ht include, among others: contemplating the pros and cons of knee replacement surgery, or choosing a novel prosthetic, an innovative surgical method, or an inventive form of therapy. In this way, not only are ideas about knee arthroplasty communicated, bu t modeled ideas, values, norms and notions concerning social conduct are also diffused throughout the network. Technology has thus expanded the range of social influence and transformed the social diffusion process as it was originally conceptualized (Roge rs, 1962). Models, in the form of opinion leaders, motivate, inform, exemplify and legitimate innovations, and serve as advocates for innovations by encouraging others in a group or network to adopt them (Bandura, 1986). People are initially reluctant to


61 u ndergo unfamiliar procedures that involve costs and risks until they assess the advantages or disadvantages that models provide. Social Power Leadership opinion or otherwise cannot occur in a vacuum. Heterogeneous individuals must be present in groups or networks to allow for interaction between and among them. From the outset in the study of group dynamics, it has been assumed that morale, group effectiveness, and leadership are all intimately related to one another (Cartwright & Zander, 1968). Despite th e complexity of these relations, the nature of a Leadership consists of such actions by group members that move the group toward its goals, improve interaction among membe rs, build cohesiveness, and make resources available to the group or network. As specific members perform these functions, they may be seen as leaders, which translates into social influence, which then translates into social power (Cartwright & Zander, 19 68). Any realistic description of group, organization, alliance, federation, or network dynamics must recognize that leadership inevitably involves the ability to influence other people in some way (Dahl, 1957). Even in highly informal, voluntary networks leaders are recognized by their ability to affect the course of events within the association. It follows then, that if social power consists of the ability to influence other people by whatever means, leadership clearly involves the use of power (Cartwr ight, 1959; Lasswell & Kaplan, 1950). Although there are undoubtedly many possible bases of social power, French and Raven (1968, p. 263) identify five which they deem to be common and most important: Reward wards for B


62 Coercive Legitimate based on perception that A has the right to prescribe behavior for B Referent Expert based on the perception that A h as some special knowledge or expertise Although referent power could overlap here in that B may well desire to identify operative healthy state, most relevant to this research is expert power, knowledge or perception that B be contemplating it. For expert influence to occur, it is necessary for B to think that A really knows about knee arthroplasty and is be ing honest in imparting information about it. Defining Opinion Leadership As the internet continues to evolve as a milieu for social interaction, the operational definition s. Earlier research conceived of opinion leadership as an individual (Shah & Scheufele, 2006). Katz (1957) explained that an opinion leader is a person who exert s influence on the opinions of others, and can be distinguished from others via three criteria: 1) who one is the personification of certain values; 2) what one knows competence; and 3) whom one knows be f ound on every level of society and presumably, therefore, are very much like the concept while updating his theories on diffusion, Rogers (2003) identified six generali zable characteristics of opinion leaders that differentiated them from their followers: 1) greater exposure to mass media; 2) more cosmopolite; 3) greater contact


63 with change agents; 4) greater social participation; 5) higher socioeconomic status; 6) more opinion leaders are more innovative, but when norms do not, opinion leaders are not especially innovative (pp. 316 318). While operational definitions of opinion leadership mea surement have varied among studies, some variation of the four key methods utilized in earlier research ld be best identified This approach came to be known as the sociometric method, and emphasizes the associational element of this individual disposition. Group members are asked whom they go to for advice or information about a given issue (Who is your leader?). Without a full census of the population under study, however, gaps in the network structure might lead to inaccurate assessments of opinion leadership. This meth od is most applicable to 1962, p. 438). (Although Rogers [1995] notes it is possible to locate sociometric leaders through snowball sampling.) In order to address the problems identified in the sociometric method for studies with larger populations, key informants can be interviewed and asked to identify opinion leaders (Who are the leade rs in this system?) (Weimann, 1994). Self group or network (Are you a leader in this system?). This method is open to challenge


64 from a number of perspectives, primarily due to de pendence on the accuracy of report. Opinion leaders can also be identified through observation. In information. Before the proliferation of online fo rums where all active members leave recoverable text based trails through posts the shortcoming of this method was self and required a relatively longer time t The scope of the observation method has been widened by the potential the internet provides. Opinion leader identification techniques differ on their ability to capitalize on each who on e is, what one knows, whom one knows dimensions. For example, self whereas sociometric methods may provide a better measure of social position. Thus, one factor that influences the se lection of an identification method is the importance of each of these dimensions to the research project (Valente & Pumpuang, 2007). Opinion Leadership in a New Age As rapidly as communications technology has changed in recent years, so have the ways in w hich people form social networks, disseminate information, and exert influence within those networks. Since a chief component of being an opinion leader is communicating with others, changes in the communication environment have noteworthy implications for opinion leadership. According to McQuail (2000), the internet, the most prolific representative of new media, can be seen as a virtual substitute for interpersonal communication, interactive play, information search and collective participation. Research focusing on the emotional nature of social computer


65 mediated communication (e.g., Lea & Spears, 1992; Parks & Floyd, 1996; Walther, 1995; Walther, Anderson, & Park, 1994), based on principles in social cognition and interpersonal relationship development f rom social psychology, suggest that given enough time, individuals can create fully formed impressions of others based solely on the linguistic content of written electronic messages. Despite limited social cues, social resources such as emotional support, leadership, companionship, a sense of belonging, and group identity are visibly exchanged online between individuals who do not know each other in the offline environment (Haythornthwaite, 2002). In an environment that proffers reduced social cues, other psychological comparisons, such as an inclusive mind set (Blanton, 2001) or shared group identity (Brewer & Webber, 1991) are more important to enhancing communication effects than is structural similarity the shared social characteristics such as gender, age, ethnicity, physical similarity, occupation or socioeconomic status (Wellman, 1997; Kollock & Smith, 1999). Most vital to enhancing communication in an online health community, however, is experiential proximity the knowledge or perception that those with whom one engages have a certain degree of experience or understanding of their own situation (Maloney Krichmar & Preece, 2005; Tanis, 2008). Weak ties and expertise are crucial because opinion leaders often base their remarks on acquired knowledge or firsthand experience in their area of expertise, thus their advice reduces perceived risk leaders provide both favorable and unfavorable information adds to their credi bility (Jin et al., 2002).


66 Several recent studies have attempted to determine how opinion leadership is based political dis cussion, Rhee, Kim, and Kim (2007) identified online opinion leaders as those who received more postings and produced more positive reactions than the negative reactions in the portal based discussion plaza. Online opinion leaders were found to be 11% of t he sample, predominantly male, educated, and slightly older than the online general public. The researchers also found these leaders to be more competent in internet communication than the general public in the sample. In a marketing study utilizing both quantitative and qualitative data, J.O. Brown et al. (2007) referred to online information exchange as word of mouth theory and found that respondents evaluated the credibility of information both in terms of the individual source and the website used, whi ch ultimately affected the value of that information. With websites themselves acting as social proxies for tie strength and homophily identification, leadership in this context was dominated by the website; respondents commonly mentioned themes which disp layed variations of social affiliation with websites whose content, rather than the characteristics of individual members, demonstrated similarity of interests with the user. Here, content refers to the actual textual content (i.e., the information content ) of the website, rather than who actually provides that information (i.e., the individual users). Researching social interaction in an online forum focused on knee replacement, Maloney Krichmar and Preece (2005) discovered that participation patterns eme rged centering around a group of key community members. Some of these key members


67 participated in leadership capacities for long periods of time, some had high participation rates for sporadic and shorter periods of time, and some had been active participa nts but left and came back to help new members. These patterns indicated there were newly injured individuals frequently joining the community who needed information, advice, and support, and there were long term members the opinion leaders who possessed t he knowledge and experience to meet these needs and desired to do so. The researchers concluded that the success and vitality of this community was made possible by a stable core of long term members who provided guidance to newer members. Whereas the cu rrent study utilized Social Network Analysis, an online questionnaire, and online interviews in a health context to identify opinion leaders, Rhee et al. (2007) and Brown et al. (2007) determined leaders by numerically valuing political postings and websit e content, respectively. Rhee et al. identified leaders through observation according to predetermined cri teria (high number of postings and high number of positive responses to postings). Leadership emerged for Brown et al. through participant consensus a fter analysis of interviews and postings. Similar in subject matter to research conducted for this study, Maloney Krichmar and Preece (2005) also looked at knee health, yet they sought to help health care professionals find ways to build and maintain healt h related support forums. These researchers identified opinion leaders via observation and subsequent categorization of participants according to predetermined external criteria: group m embership role analysis developed by Benne and Sheats (1948).


68 Social Network Analysis More recent studies of interaction on the internet have explored opinion leadership in online contexts utilizing Social Network Analysis (SNA) (Hanneman & Riddle, 2005; Lewin, 1936; Moreno, 1934; Scott, 2000; Wellman, 1983; Wellman & Berko witz, 1988), a theoretical perspective and set of methods for analyzing social structure. While SNA is rooted in psychology, social psychology and anthropology, it is well known in communication studies (Haythorthwaite, personal communication, July, 2010) and is being employed more frequently in mass communic ation research (Caiani & Wagemann, 2009; Clouston, Verdery, Amin & Gauthier, 2009; M. G. McGrath, 2009; Prell, 2003). As flows of information and opinion leadership are inarguable mechanisms of social s tructure, these concepts can be analyzed in terms of the relational ties and the patterns of ties that link members of a social structure in this research, a social structure in the form of a community created and maintained in virtual space. One of the mo st potent ideas in the social sciences is the notion that individuals are embedded in thick webs of social relations and interactions. Social network theory and analysis provides answers to questions that have occupied social philosophy since the time of P lato, namely, the problem of social order: how autonomous individuals combine to create enduring, functioning societies (Borgatti, Mehra, Brass, & Labianca, 2009). SNA also has the potential to provide explanations for numerous social phenomena, from indiv idual creativity to corporate profitability (McCarty & Bernard, 2003). SNA focuses on patterns of relations among actors; any theoretically meaningful unit of analysis may be treated as actors: individuals, groups, organizations, communities, states, or c ountries. (Streeter & Gillespie, 1992; Wasserman & Faust,


69 1994; Wellman, 1983). Social network analysts seek to describe networks of relations, tease out prominent patterns in such networks, trace flows of information (and other resources) through them, an d discover what effects these relations and networks have on units of analysis (Garton, Haythornthwaite, & Wellman, 1997). The use of computer software (UCINET6 in the current study) allows the researcher to generate various social network maps visually d epicting communicative relationships among members of a social system, providing realistic rather than probabilistic results. These maps illustrate not only the direction of relationships, but their strength as well. The visualizations enable researchers t o analyze social structure; opinion leaders in a social system can also be identified via maps. Depending on research goals, the focus of visualization and social mapping can be on identifying opinion leaders through interpersonal communication, excluding the influence of external media, a major variable in traditional studies of opinion leadership ( D. K. Kim, 2008). Common measures utilized in SNA are centrality, distance, degree, density. In a study of logic based programming, M. G. McGrath (2009 ) explore d a website focused on consumer generated content formed around an Australian Football League. analyzed several hundred posts from what he determined to be the 20 most signif icant contributors in this social network. Attitude scores were assigned to posts among actors and averaged for each pair of actors. The resulting sociogram and matrix demonstrated that; a) participants were divided into two distinct cliques, connected by a bridge anchored by two particular actors who attracted significant direct negative comment, and; b) a clique of like minded individuals formed around two other participants.


70 Mathematical procedures produced measures of centrality indicators of leadership and influence for each participant. After revising the analysis to exclude negative comments, McGrath established that two actors in the clique of like minded individuals were most influential, thus any external attempts to sway the thinking of members of this conclusion was that in the bigger picture of consumer internet use, organizations are recognizing that it may prove fruitless to compete with peer networks and inst ead must work with (and within) them. To do this, organizations must understand relevant network structures; particularly influence hierarchies and key influence actors. Utilizing SNA and UCINET6 ( Borgatti Everett, & Freeman, 1992, 1996, 2006) and looki ng for opinion leaders in a study for an international business firm X. Zhang and Dong (2008 ), investigated an online real estate forum with over 5,000 members. These researchers found that because of largely anonymous communication and mutual weak ties a mong participants, the key informant and self identification methods of identifying opinion leaders did not work in this online context. Through observation of postings and subsequent measures of centrality, the researchers identified 20 opinion leaders th e most active in providing real estate advice. Citing past success in the application of diffusion theory, the researchers recommended the firm target these 20 and incentivize them through firm related empowerment and experience. In a diffusion study in I ndia, D. K. Kim, Chitnis, Vasanti, and Singhal (2007) explored the communication networks of 225 farmers in 14 villages, identifying opinion identify opinion leaders th rough SNA and analyze the attributes of opinion leaders in the


71 diffusion of agricultural related information delivered via internet kiosks, or e Choupal centers, placed in villages throughout Central India. Between 2000 and 2007, the number of e Choupal ce nters increased 361%, expanding from 1,800 covering 8,000 villages to 6,500 covering 38,000 villages. The e Choupal consists of a computer with multimedia features connected to the internet by a dial up or via a VSAT connection. The e Choupal center is ins talled in the house of a community member trained to use the technology, who then provides farmers with information about current market prices, provisions to sell crops directly to buyers, and up to date information on weather, farming practices and insur mass media sources of information were available via the e making during adoption processes was determined more by their personal contacts with opinion leaders, even though mass media functioned as a source of new knowledge. While measures of centrality can be used to identify opinion leaders, D. K. Kim et al. (2007) argue in the e Choupal study that opinion leaders who are the key players in accelerating diffusion are not just those who have higher centrality measures, but also those who are strategically positioned to reach everyone in the network. As expected, three of the four e Choupal trained community members in the initial 14 villages were found to be opinion leaders sources, a substantial number of other farmers who were not opinion leaders were cited as information sources as well inno vations were also introduced to farmers by non opinion leader peers, these farmers


72 still felt the need to confirm adoption decisions with opinion leaders and not with those who were the initial source of information about the innovation. While McGrath (2 009), X. Zhang and Dong (2008), and D. K. Kim et al. (2007) all utilized SNA to seek out opinion leaders, McGrath, and X. Zhang and Dong sought to advise organizations, and D. K. Kim et al. conducted only face to face interviews with targeted respondents. While suitable for those researchers, these distinctions differentiate those studies from the study proposed herein, in both aim and method. Social Network Analysis in Mass Communication While not exploring opinion leadership explicitly, a number of stud ies conducted in mass communication have utilized SNA as perspective and method (Haythor nthwaite, 2000). Although it has been used in other disciplines, only recently have scholars employed SNA to understand relationships in mass communication environments such as, for example, social network sites (SNS) like Myspace, Facebook, Cyworld, and Bebo. These networks are populated by millions of users, a large number of whom have incorporated SNSs into their daily practices (Boyd & Ellison, 200 8 ). In what they t ermed et al. (2009) created an original associational network dataset from Facebook users of an undergraduate university (enrollment 16,000). These researchers saved pages for each person in their random sample and ca chosen to deny permission).This data indicated ties and attributes of those connected by gender, faculty/school, class levels, country of origin, and graduating year. Data also included information about organizational affiliations (workplaces and regions) to which file


73 information, with matrices constructed to indicate if two individuals in the sample shared the same categorical qualities. Among findings, Clouston et al. (2009) an actor had in their sample was 176, and tha t men were more connected than women on campus, though they found no gender differences in off campus ties. Year of expected graduation was the strongest correlate of being connected, while gender was weakest. What is noteworthy about this study is that so cial network sites can potentially provide considerable network data in a cost effective and efficient manner. Studies that explore associational ties of individuals in large social groups often rely on resource intensive procedures to gather associational data, such as multiple interviews and/or with even with limited resources, large and fairly complete social network data can be gathered which can illuminate the soc ial interactions of individuals in a virtual society; a less Caiani and Wagemann (2009) applied SNA to a study of communication networks within the Italian and German extreme ri ght. Using Web links between organizational websites as proxies, these researchers found that extremist groups increasingly use a nd abuse the internet for propaganda and recruitment, a s well as for internal communication. Using the most common measures off ered by SNA distance, degree, density and centralization the investigators were able to determine whether these organizations operated cohesively or were disjointed. They found the Italian network to be fragmented, highly diversified a nd difficult to coor dinate across entities, and the German network denser and more concentrated on a few central actors ( resembling


74 star structure characteristics of social networks ). T reat ing the Web links as indicators of closeness, instruments for reciprocal help in attain ing public recognition, and as potential means of coordination the study concentrated on connections between the micro and macro dimensions of this issue. By focus ing on the level of individual organizations, examining which types of organizations occupy a central positio n and which are more peripheral, the research explore d the fo rmation of specific alliances of communications between the vario us extremist groups, looked at their composition (e.g. their homogeneity or heterogeneity) and at the relations b etween and among them ; conflict and consensus were factors as were how s egmented or centralized these virtual networks appeared to be In this context, Caiani and Wagemann found that the structural properties of these networks provide the w ebsites with op portunities to become substitutes for empirical organizational structures. SNA researchers focus on the structure of social networks and see this structure as an indicator that social capital exists. Although SNA scholars agree that structure is an indica tor of social capital, differences in perspective exist as to which structures act as the best indicators. Prell (2003) investigated the social aspects of a community network by concentrating on how the representatives of local non profit youth agencies an d community leaders communicate with one another, share resources, and locate trust in each other. Concentrating on the development of one component of a social capital, relations wit hin a community section of non profit youth agency representatives, faculty from a local private research university, city government


75 employees, and administrators from the local school districts and utilizing SNA and UCINET Prell measured the social capital concepts of frequency of communication, trust and exchange of resources. Interested in determining centrality as it relates to network structu re and social capital in this context, the study evaluated both in degree betweenness ( measuring where a particular actor lies between other actors in a network) centrality. A ctors who receive information from m any sources are considered prestigious a ctors with high betweenness scores often serve as gatekeepers and resource brokers (Diani 2003 ) show ed that ma ny of the same actors dominated the network in both in measures of centrality an d, interestingly, tha t most of the popular actors were also employee s of government agencies the same agencies that circulate fund ing resources among non profits organizations. Thus, funders in this network seem to hold a large amount of social capital. Th e data also indicated that in degree scor es for social interaction had significant positive correlations with trust and resource exchange an actor's popularity (high in degree ) coincides with othe rs' nominating that actor as one who gives funding and is tr ustworthy. profit organizations have traditiona ll y lacked both funds and staff, they tend to focus energies on serving clients, not on connecting with actors outside the ir organizational settings. An exception, however are actors whose organizations provide needed resources such as funding. More specifically, government agency employees were the most po pular and powerful actors in this network Prell argues that these results also show that in certain scenarios, in deg ree and betweeness centrality are virtually synonymous with one another


76 Clouston et al. (2009) Caiani and Wagemann (2009) and Prell (2003) all utilized SNA to analyze communication patterns, however, unlike the current study, SNA was used as the primary method of investigation and several mathematical procedures were employed to obtain results. These studies were also conducted in contexts notably different than the health related environment herein, and none specifically investigated opinion leadership, objectives. These studies however, demonstrate SNA social interaction. Two Step to Multi Step Despite criticism by later researchers directed at its oversim plification, the two step flow hypothesis has remained relevant to communication study (Weimann, 1994). Given the literature reviewed for this study, it is apparent that the process of influence is more complex than opinion leaders simply gleaning informat ion from mass media messages and passing their opinions on to passive followers. People who influence others are themselves influenced by others in the same areas of interest, resulting in an exchange. Thus, opinion leaders are both disseminators and recei vers of influence (Weimann et al., 2007). The proliferation of digital media technologies and the diffusion of the internet have now radically expanded access to readily available sources for information of all types. Much of the information once obtained secondhand from friends and acquaintances the traditional two step flow is now easily found through the internet and world wide web utilizing any number of digital communications protractions (Case et al., 2004). This, and accepting that opinion leadership does not occur in a vacuum, generates a different account for human interaction; a more accurate depiction of


77 modern communication flow, then, would include the recognition of a multi step process. With few exceptions (some noted above), the concept of opinion leadership has been researched within the context of traditional forms of community and social interaction (Nisbet & Kotcher, 2009). The customary view holds that individuals sharing geographical space engage in primarily face to face communication discussing news, society, personal interaction is no longer dominated by physical proximity, relatively strong interpersonal ties, and face to face conversation, but rath er is shared with geography spanning digital interactions that involve many loose ties (Boase, Horrigan, Wellman, & Rainie, 2006). The increasing use of internet sources for information can be seen as a further mutation of the two step flow hypothesis. It is obvious the internet and world wide web have changed the nature of information seeking, as the internet is often the first choice to inquire about a host of topics, and serves as a conduit for discussing those topics with other internet users (S. Fox 2010; Hesse et al., 2010; Kaye & Johnson, 2003). Rains (2007) reports that those who used the internet for health information rated information they obtained from family and friends to be significantly less useful than information obtained from the inter net, while those who did not use the internet rated information they obtained from family and friends to be more useful. (p. 670) Although knee, hip and joint replacement are becoming more routine, it could be that circumstances surrounding knee arthropla sty or its potential are new enough to where involved individuals are hard pressed to find someone in their immediate circle


78 with knowledge sufficient in this context to be considered an opinion leader. This could be also be an aspect of a broader trend th at A. Shapiro and R. Shapiro (1999) refer to bypass experts in its quest for information, products, and services. Perhaps, faced with the wide array of information sourc es now available, the average person is more interested in gathering facts and less interested in being convinced or persuaded by in this case, the medical establishment. The internet and world wide web, in their accessibility, anonym ity and potential interpersonal authoritativeness, may now act as substitutes for the classic two step model, supplanting interpersonal networks traditionally based on the dynamics of social and physical proximity. Case et al. (2004) contend that the rise of technology coupled with the decline of the American nuclear family has resulted in less reliance on family members for information, and may account for some of these changes in source preferences. Another prime variable in past studies falls under the collective leader attributes. Although exposure to mass media, higher socioeconomic status, etc. [see above]) are widely reported as important variables in the search for opinion l eaders, and while these normative and overarching attributes provide a researcher with a general sense of who opinion leaders might be, it is difficult to actually identify opinion leaders in a social system employing these general attributes because all s ocial systems have their own particularity and uniqueness. This research determines how opinion leaders can be identified among people in this system, rather than relying on who opinion leaders are based on traditional generalized attributes.


79 Antecedent s of Effective Communication Scholars out of the Chicago School were instrumental in early sociological research, with studies concentrating on individual differences that characterized the people who composed audiences for mass communication and how those differences shaped their psychological a nd overt responses to the media (Lowery & DeFleur, 1995). Later, Lewin (1936, 1948) initiated a shift to the study of group communication; how social category memberships among members of an audience influenced patt erns of attention and response. represents the first clear and intensive focus on social relationships and their role in mass communication processes. Ties between and among people were seen as mo re important factors, rather than the structure of messages, perceived characteristics of sources, or the psychological makeup of the message receiver (Lowery & DeFleur, 1995). Over the years however, communications research has consistently advanced knowl perceptions of sources affect interpersonal communication and vice versa, a critically important component in the human communication process (L. L. McCroskey, J. C. McCroskey, & impression of the source. Perceptions of the source, therefore, determine perceptions of the message McCroskey & Richmond, 1996, p. 104). Virtual communities are sources of and for information and are typically emergent, they arise as a natural consequence of people coming together for any number of reasons; e.g., to discuss a common hobby, medical affliction, personal experience, or to develop relationships around a lifestyle or co nsumer product (Ridings et al., 2002). The rapid growth of virtual communities on the internet and accompanying surge in interest


80 by researchers (Hine, 2000; Kozinets, 2002; Preece & Ghozati, 2001; Rafaeli & Raban, 2005; Wellman, 1997; Wellman & Gulia, 199 9) raises questions as to what encourages members to interact in virtual communities; what makes them initiate contact, attend and respond to messages, revisit, and in many cases, remain in these communities. This study suggests that trust, credibility, an d homophily (similarity) are elements generated by individual perception that help promote voluntary online cooperation among strangers in virtual communities. Differences between online and offline communication are important because member trust in elec tronic communities must accordingly be made in most cases on the basis of written communication only. Face to face interaction is fundamentally different from written communication in that it allows for an exchange of verbal and non verbal information (Hil tz, 1984; Rheingold, 1993). Additional meaning perceived in face to face communication and carried via inflections of voice, gestures, dress, tone, posture, and other indicators, is missing in written communication so that messages remain open to multiple interpretations (Sproull & Kiesler, 1991). Trust While trust and relationships borne out of it have been topics of research in many disciplines since the 1950s, it is rooted in the concept of ethos trust in, or credibility of, the source of a message. Etho s, along with pathos (emotional appeal), and logos (logical appeal), are components of what the ancient Greeks termed rhetoric; the art of effective communication. Streams of research on trust can be found in the fields of philosophy, sociology, psychol ogy, management, marketing, ergonomics, industrial psychology, and e commerce (Corritore, Kracher, & Wiedenbeck, 2003) and when considering these various disciplines together, the literature on trust is


81 quite extensive. While trust has been studied in a va riety of disciplines, each has produced its own concepts, definitions and findings. Even within a given field, there is often lack of agreement on its nature (Lewicki & Bunker, 1995). Reasons for multiple definitions of trust in the literature are likely t wofold. First, trust is an abstract concept and is often used interchangeably with related concepts such as credibility, reliability, or confidence (Y. Wang & Emurian, 2005). To define the term and differentiate the distinction between trust and its relate d concepts has proved challenging for researchers. Second, trust is a multi faceted concept that incorporates cognitive, emotional, and behavioral dimensions (D. Lewis & Weigert, 1985). In a critical analysis of trustworthiness as a component of security, Nissenbaum (2001) stated, variety of objects. One can trust (or distrust) persons, institutions, governments, information, deities, physical things, systems, and of uniform premises by which to define trust and its manifestations. The meaning of trust is dependent upon the situation in which it is being considered, another factor contributing to the complexity of a definition (D. Lewis & Weigert, 1985; Luhmann, 1979). In a virtual community, one can be conversing with one or two other individuals, thus trust can be understood in the context of interpersonal relationships, i.e. trust between people (Rotter, 1967). At the same t ime, because these same posts are seen by the general audience, trust is at the generalized, collective level, and notions of interpersonal trust are applied to collective entities (Jarvenpaa, Knoll, & Leidner, 1998), including networks like these (Ridings et al., 2002). Others, however, contend it is too difficult to assess the trustworthiness of remote entities,


82 because computerized communication media are increasingly removing familiar styles of interaction (Jsang, Ismail, & Boyd, 2007). Jsang et al. a rgue that physical encounter and other traditional forms of communication allow people to assess a much wider range of cues related to trust than is currently possible through computer mediated communication. While again there are many definitions of trus t, only a few will be noted here. be said to have trust in the occurrence of an event if he expects its occurrence and his expectation leads to behavior which he perceiv es to have greater negative motivational consequences if the expectation is not confirmed than positive motivational ancy held by an individual or group that the word, promise, verbal or written statement of another individual or group can be psychological state of the trustor. Mayer, D avis, and Schoorman (1995) defined trust as expectation that the other will perform a particular action important to the trustor, irrespective of the ability to moni which incorporates risk. Mayer et al. argue there is no need for trust if there is no risk in a situation. can be defined as dimensions of trust: benevolence (you care about me and take an interest in my well


83 being and goals) and competence (you have relevant expertise and can be depended upon to know what y ou are talking about). vulnerability must also be present in any definition, offering that trust is an overriding erabilities. Vulnerability implies that the trustor must be exposed in some way. In the online environment, the trustor could be exposed through a lack of knowledge or expertise, or an inability to acquire something without the assistance of others. Trust encompasses the perception that one has vulnerabilities and that those vulnerabilities could be breached or capitalized upon (Sabel, 1993). Some e commerce researchers have defined trust as a belief in an attribute of the trustee (McKnight, Choudhury, & Kacmar, 2002), whereas others have defined trust as a trustor based willingness to believe the trustee (Fung & Lee, 1999). Much scholarly research on trust has supported the assertion that the concept is multidimensional, although studies vary on what co mprises those dimensions. Many researchers contend that trust consists of three distinct dimensions: ability, benevolence, and integrity (Blau,1964; Butler, 1991; Giffin, 1967; Mayer et al., 1995), although often these are intertwined (Y. Wang & Emurian, 2 005). Each dimension has been shown to be relevant in the case of online communication (Jarvenpaa et al., 1998). Ability corresponds to a belief or perception held by the trustor in the skills or competencies of others which enable those others to have in fluence in a certain area (Ridings et al., 2002). In the environment of a virtual community, participants perceive


84 the community or on the belief 2008). This applies to virtual communities because they are almost always centered around a specific mutual interest, hobby, life event, or occupation, and perceptions about the abilities of others with respect to mutual concerns are important to interaction (Ridings et al., 2002). Benevolence is the expectation that others (trustees) will have a positive orientation or desire to do good to and/or for the trustor (Wu & Tsang, 2008). In a virtual community, trustees respond to requests for information or make general posts with appropriate advice, support, discussion and such, contributing to the ongoing interactions with intention to help support, and care for others. Benevolence is important here, particularly in a health environment, because without positive reciprocation a nascent network would not be in existence long enough to become a virtual community. Benevolence is initially diff icult to ascertain, but in a health related virtual community, there is generally no material transaction taking place so someone looking for information has time to build trust in others providing it ( J. Boyd, 2003). Deutsch (1958) considered integrity t o be a component of personal responsibility in arguing that personal integrity is related to such personal characteristics as the strength of internalized values with regard to responsibility, the ability to prevent or resolve conflict in responsibilities, and the ability to realize the goals of others as goals of one's own (p. 268). In the online environment, Jarvenpaa et al. (1998) assert that integrity is the expectation that others will act in accordance with socially accepted


85 standards of honesty or a set of principles (both tacit and stated) that all within the community ostensibly accept. Research has identified several additional antecedent factors conducive to trust building. Disposition to trust, responsiveness, the disclosure of personal informat ion, a desire to exchange information, reputation, and identification based trust, all have bearing on the establishment and evolution of trust, and all are applicable in online environments. Disposition to trust is defined as a general propensity based o n socialization to depend on others (McKnight et al., 2002), and has been found to be related to initial trust (McKnight, Cummings, & Chervany, 1998). If one is willing to trust others in general, then this same willingness should apply to others in the vi rtual community. Disposition to trust may be especially effective when interacting parties are as yet unfamiliar with one another (Mayer et al., 1995; Rotter, 1971), as might initially be the case in a virtual community where any and all members can reply to a post. Initial trust (McKnight et al., 1998) refers to trust in an unfamiliar trustee, a relationship in which actors do not yet have credible, meaningful information about, or affective bonds with, each other (Bigley & Pearce, 1998). Information may b e deemed credible after parties have interacted for some length of time. Initial trust also reflects assumptions about how trust forms. The cognitive based literature on trust posits that trusting beliefs may form quickly (before actors have meaningful inf ormation about each other) via perceptions of social categorization, reputation, disposition, institutional roles and structures, or out of the need to immediately cooperate on a task (McKnight et al., 1998; Meyerson, Weick, & Kramer, 1996). The knowledge based trust literature (Blau 1964; Deutsch, 1958),


86 however, indicates that trust develops gradually through experiential social exchange (Lewicki & Bunker, 1995; D. Shapiro, Sheppard, & Cheraskin, 1992). A member who posts messages in a virtual community m ost often expects responsiveness some type of reasonably timely response. If no responses appear, trust in others will not develop (Ridings et al., 2002). If others respond quickly and often, it may be that they have the skills and competencies that afford them to exchange accurate and helpful information, which increases positive perceptions of their abilities. Greater responsiveness indicates a willingness to help other community members, which also increases the reciprocal nature of the community itself. Responsiveness has been associated with increased perception of cooperative intentions (Gefen & Ridings, 2002) and indicates both benevolence and integrity as it illustrates behavior according to accepted social rules and adherence to norms. People have been found to be comfortable disclosing personal information via computer mediated interaction (Sproull & Kiesler, 1991). If members post personal information about themselves, they make themselves appear to be more than just strangers, and are demonstrat ing that they trust others with sensitive information. Earlier research on social behavior indicates that by confiding and making oneself more vulnerable in the process, people induce others to trust them more (Blau, 1964; Crosby, Evans, & Cowles, 1990). B y disclosing gender, age or perhaps personal problems, or by posting otherwise sensitive information, members in a virtual community are behaving in a trusting manner and become less like strangers and more like acquaintances or friends (Ridings et al., 20 02). Additionally, since many communities form around a common interest or topic, disclosing personal information about oneself


87 ability about the topic. For example, i n a virtual community focused around knee replacement, revealing that one has had double knee replacement or is a transitional specific topics. The development of benevol ence and integrity are also influenced by divulging personal information. Knowing more about other members makes it easier to shape beliefs regarding their standards and principles, contributing to increased trust; trust is enhanced when the trustee behave s as expected (Luhmann, 1979; Blau, 1964). Trust determines, at least to some degree, the nature of interpersonal relationships (Blau, 1964; Jarvenpaa et al., 1998), and thus influences the desire to exchange information. In a trusting environment people a re more inclined to help others about providing help (Blau, 1964; Luhmann, 1979). When trust exists among individuals, they are more willing to partake in shared activity In a virtual community environment, this shared activity is in the form of cooperative information exchange, and is usually text based (Ridings et al., 2002). The concept of reputation is closely linked to trust, but there is a clear and important diff erence (Jsang et al., 2007). Reputation is the estimation in which a person or thing is commonly held whether favorable, unfavorable, or indifferent. This definition corresponds with the view of social network researchers that reputation is a quality deri ved from the underlying social network, visible to all members (Freeman, 1979; Marsden & Lin, 1982). As an example, the difference between trust and reputation can be illustrated by the following plausible statements:


88 (1) I trust you because of your good r eputation (2) I trust you despite your bad reputation Assuming the two statements relate to identical transactions, Statement 1 reflects Statement 2 signals that the trusto r has some private knowledge about the trustee (e.g. through direct experience or intimate relationship) and these factors override any reputation the person in question might have (Jsang et al., 2007, p. 620). These distinct observations reveal that trus t ultimately is a personal and subjective phenomenon and is based on various factors or evidence, and that some of those carry more weight than others. Personal experience typically carries more weight than secondhand referrals or reputation, but in the ab sence of such, trust often has to be based on referrals from others (Xiong & Liu, 2005). Last in this truncated list of factors antecedent to trust building is identification based trust, a more relationally based concept extending beyond that which can b e explained by contextual analysis. The basis for this perspective is consistent with social identity theory (Tajfel, 1974, 1978; Tajfel & Turner, 1981), which emphasizes the perception of oneness with, or belongingness to, another. Research in social iden tity theory suggests that individuals define themselves partly in terms of salient group memberships, such as organizational affiliation, religion, shared interests, etc. Such a shared sense of identity among actors may promote a deeper understanding of ot thoughts and actions and thereby enhance communication. In addition, shared identity can create a sense of empathy and conc ern for outcomes pertaining to others (Naquin


89 & Paulson, 2003). A shared sense of identity essentially yields more trusting rel ationships (Lewicki & Bunker, 1995). In an investigation of online group dynamics, Sassenberg (2002) made a distinction between common bond and common identity groups. Commo n bond groups were found to be grounded in personal bonds and interpersonal at traction among group members, whereas common identity groups share an identity and members are attracted to this identity without that attraction being grounded in interpersonal relations. Sassenberg also found that while common bond groups show greater in terpersonal attraction, common identity groups exhibit greater collective conformity to group norms and stronger social identity. Researchers agree that trust provides a mechanism to decrease complexity in a complex world by reducing the number of options one has to consider in a given situation (Luhmann, 1979; D. Lewis & Weigert, 1985). Trust can also be viewed as a kind of social capital that makes coordination and cooperation between people possible (Corritore et al., 2003). Since trust can mitigate ris k, fear and complexity in the offline environment, it is likely to do the same in the online environment. Likewise, since trust is social capital that can create cooperation and coordination in the offline environment, it too is likely do the same online. Credibility Credibility has also been conceptualized and studied utilizing a variety of methods sub dimensions source, message, and medium credibility studies show conside rable overlap between and among these dimensions (Chaffee, 1986; Kiousis, 2001; Slater & Rounder, 1996; R. G. Smith, 1978). Past credibility research suggests that a range of


90 theoretical dimensions and empirical measures may be relevant to assessing credib ility. These dimensions include safety, qualification, and dynamism (Berlo, Lemert, & Mertz, 1969) authoritativeness, sociability, character, competence, composure, and extroversion (J. C. McCroskey, 1966; J. C. McCroskey & Young, 1981). The relationship b etween receiver and source has also been identified as an important factor in determining the degree of credibility accorded a source (Aune & Kikuchi, 1993; O'Keefe, 1990). In addition, perception and measurement of source credibility may differ depending upon the type of source being evaluated and the context in which evaluation occurs (Cronkhite & Liska, 1976; Delia, 1976; Gass & Seiter, 2002; Gunther, 1988; Stamme & Dube, 1994). Traditionally, credibility research focuses on questions of what makes mess age sources believable and persuasive. Although notions of ethos date back to ancient Greece, an early and important theoretical formulation divided source credibility into two dimensions: expertise and trustworthiness (Hovland, Janis, & Kelly, 1953). For these researchers, expertise is defined as a communicator's qualifications and/or ability to know the truth about a topic. Trustworthiness relates to a judgment about the communicator's motivation either to tell the truth about a topic, or to bias informat ion for self serving motives (such as commercial gain). J. C. McCroskey and Richmond (1996) also divided source credibility into two dimensions, yet for them, competence replaced expertise, while trustworthiness was defined in terms of character. In defin ing competence, these researchers argued that a completely incompetent to completely competent. The perception was mediated,


91 however, by how competent the receiver perceiv ed themselves to be if one deemed likely influence the receiver. The researchers also contend that competence is usually s credibility (p. 107). because people discount the validity and honesty of messages from sources perceived to be untrustworthy. Although a source may be perceived to be compete nt on a topic, credibility may not follow if a receiver feels the source is not generally honest. Even if messages sound plausible, if a source is thought to be untrustworthy a receiver then reliability of the message (J. C. McCroskey & Richmond, 1996). Both expertise and trustworthiness are reflected in online communication. Studies show that users rely on experts to answer questions, get advice or find useful information (Wasko & Faraj, 20 appear informed and knowledgeable, perhaps even well written. In addition, these sources must be considered trustworthy. Trustworthiness is enhanced when the source appears embedded in the group and de dicated to its success (Tanis, 2008). Many virtual communities emphasize this aspect by providing additional information about ish an individual with exceptional tenure in the community. Even if these attributes are not directly available, users can usually determine how long someone has been in the community based on the time stamps associated with posts (Huffaker, 2009).


92 In qua litative accounts of face to face support groups, participants have reported that knowing others in the group have faced similar problems and circumstances lends credibility to advice and support offered by those members (Balk, Tyson Rawson, & Colleti Wete l, 1993; Wright, 1997; Young & Tseng, 2008). In online environments, credibility assigned to support providers also largely derives from experiential proximity the perception that the provider has been through similar circumstances, has had similar problem s and engaged in similar behaviors, and has similar attitudes and beliefs about the condition he or she is facing although this perception emerges through text based interaction (Maloney Krichmar & Preece, 2005; Preece, 2000; Preece & Ghozati, 2001; Tanis, 2008). According to Walther (1995), the amount of time an individual spends communicating with others online provides opportunity for more messages to be exchanged, thus impressions of people are modified as the number of messages increases. As messages a ccumulate and people learn more about their conversational partners, the additional information appears to help people circumvent the lack of nonverbal cues. As a result, relationships become similar to face to face interaction, providing a mechanism for a ssessing credibility. Galegher, Sproull, and Kiesler, (1998) found that legitimacy and authority were the two most important factors in establishing credibility in electronic groups. Legitimacy or her concern is genuine and justified, and is worthy of attention; legitimacy thus prompts advice, support and information from others in the community. Authority draws on tacit claims to the appropriateness, relevancy, or sufficiency of the content mem bers share. These researchers found that participants often referred to medical or scientific research in


93 their messages as a means to establish authority. Although scientific reference is likely to be perceived as appropriate and relevant, it is not neces sarily sufficient. Information based on personal experience experiential proximity was perceived to be more relevant to other participants but was not sufficient in and of itself to establish authority when taken in isolation (Galegher et al., 1998). Whil e credibility is certainly important, Z. Wang, Walther, Pingree, & Hawkins (2008) claim credibility itself is not what differentiates influential from uninfluential counsel gleaned from online communities. Homophily is what grounds credibility and drives t he whole persuasive process in the context of online health advice (p. 365). For online peer discussion groups, homophily appears to promote message evaluation and likelihood to act on the advice. Homophily Results from earlier research of the two step flo w and personal influence demonstrated that mass media had limited effects on the American public. In the 1970s, scholars became interested in uniting what mass communication theorists and interpersonal communication scholars had in common: a shared interes t in the a mutual concern for explaining the nature of message perception and how messages are processed (Baran & Davis, 2000). Researchers began to integrate intra indivi dual, interpersonal, network and organizational, and macrosocietal phenomena in order to gain a more thorough understanding of communication processes (Berger & Chaffee, 1987). A basic principle of interpersonal communication is that source receiver sim ilarity, or homophily, increases the likelihood of communication attempts and


94 promotes communication effectiveness (J. C. McCroskey et al., 1975). While these researchers found this to be generally applicable, they claim the best example of this concept li es in the opinion leader/follower relationship most commonly observed in mass communication and diffusion research (p. 323). Homophily is the degree of perceived similarity a receiver ascribes to a message source. Homophily affects persuasion and percepti ons of otherwise unknown individuals, both offline (J. C. McCroskey et al., 1975; Suitor & Keeton, 1997; Wheeless, 1974) and online (Campbell & Wright, 2002; Wright, 2004). When people share common meanings, beliefs, and mutual understandings, communicatio n among them is more likely to be effective, and they are more likely to develop and maintain supportive relationships with others to whom they are similar on important social dimensions (Lazarsfeld & Merton, 1954). In developing an instrument to measure p erceived homophily, J. C. McCroskey et al. (1975) found four relatively uncorrelated dimensions attitude, morality, background, and appearance they believed would provide useful for researchers interested in this communication variable. The first three are still in use today for studying source receiver similarity in online contexts (J. O. Brown et al., 2007; Z. Wang et al., 2008; Wright, 2004). Earlier computer mediated communication researchers (Hine, 2000; Galston, 2000; Turkle, 1995) mention initial c oncerns over reduced social cues conditions in which the lack of nonverbal cues would lead to more antisocial communication and less satisfying relationships. Walther (1992, 1994, 1995, 1996) challenged this perspective with a series of experiments that ex tended online interaction time. Overall, Walther found in the early studies that limited online time may have created experimental


95 artifacts: diminished opportunity for satisfying relationships to develop thereby creating conditions for more antisocial beh avior. With these studies, Walther and other communication researchers (Lea & Spears, 1992; Parks & Floyd, 1996) consistently observed that over time computer mediated relationships become more similar to face to face relationships as messages accumulate b etween relational partners and users become accustomed to communicating in this environment. These studies revealed that as the number of messages among users increases, relational satisfaction also increases because people spend more time online and learn more about relational partners. Other perspectives exist however. According to the social identity/deindividuation theory of computer mediated communication (Reicher, Spears, & Postmes, 1995; Spears & Lea, 1994), it is social identity the elements of an i concepts which they derive from membership in social groups that drives online interaction. This viewpoint was echoed by researchers who found that people with similar backgrounds or similar health experiences (e.g., illness, addiction, or disability) exhibit more empathy toward each other (Maloney Krichmar & Preece, 2005; Preece, 1999, Tanis, 2008). In a comparison of demographically homophilous versus heterophilous (dissimilar) online an, Wizlenberg, Golant, and Di Minno (2005) found that members of homophilous groups were significantly more attracted and committed to their groups, reported higher cohesiveness and satisfaction, showed greater positive changes, and derived greater benefi t from group participation. Peer to peer discussion with similar others also helps reduce isolation, a circumstance not satisfied by consulting professionals or experts (Wright, 2000). The insight that


96 peers provide through online support groups is often c onsidered to be unique and facts, but other patients can tell you what it really feels like and what to expect next, in a way that only someone with personal experience can not many health care providers are involved in online support groups; most online discussion occurs between and among group members who share similar health problems, and a sense of community is important to these members. The recent rapid growth of digital media technologies has created new possibilities for people with health related concerns to engage in supportive communication with networks of individuals who may be coping with similar problems networks that would be d ifficult or impossible to form offline. This study found that several long established principles of interpersonal communication trust, credibility, and social similarity still apply. Gaps in Research While previous studies of internet interaction have e xplored forms of opinion leadership, trust, and credibility in various online contexts, research investigating these concepts concurrently in a health related online community with individuals rather than websites, institutions, or technology as targets of inquiry, is scarce. As noted, opinion leadership has been examined in diffusion of innovations, communications research, marketing, and health. With few exceptions, trust and credibility have been investigated largely in terms of how these concepts are en gaged when internet users evaluate either websites themselves, the information they offer, or institutions represented by a particular site. Phenomenological interviews have been conducted to extract meaning


97 from individuals who use Sony Walkmans, attend c hurch, visit online lesbian bars, teach Korean students, and facilitate hospital administrative policy. Yet previous studies have generally overlooked the subjective experiences of those who join online communities in search of solutions for specific heal th related conditions. Health related online communities might be different in that while participants in other types of communities might be equally motivated in levels and frequency of participation, physical pain (or the physical pain suffered by a love d one) may well be the primary factor driving participation. Previous research has shown that online health communities dedicated to a particular condition are highly supportive (Tanis, 2008), as evidenced in part by numerous, quick, and directed responses relative to overall membership. Although online health community membership is represented by people of all ages, the number of health related online communities is growing rapidly (S. Fox, 2010), a phenomenon likely due to a combination of an aging popul ation and increasing internet access. While members of all types of online communities may well be emotionally engaged in the community, considerable emotional attachment to issues that affect future quality of life are evident in online health communities This research is important in that it associates these concepts in ways previously unexplored. The Nature of Inquiry Historically, there has been heavy emphasis on quantification in science (Guba & Lincoln, 1994). This stems from early conceptions that t he path to truth and knowledge must proceed through observable and measurable phenomena a true explanation or cause of an event or social pattern can be found and tested by scientific standards of verification. As the study of human behavior emerged from i ts religious roots and shifted


98 into the domain of scientific inquiry (Babbie, 2004), it naturally fell under the control of this dominant paradigm: positivism. Positivism, which calls for measurement and quantification in order to explain cause/effect mech anisms and to generalize to larger populations while aiming to predict and control outcomes, dominated communication research until the 1980s (Lindlof & Taylor, 2002). Criticisms of positivism resulted in the rise of a postpositivist paradigm, which champi ons qualitative methods of inquiry, strains of which emphasize that claims to knowledge should preserve the subjective experiences and motivations of social actors as these actors strive to elicit meaning from experience (Lindlof & Taylor, 2002, p. 111). While positivism declares reality to be singular, objective, and independent of human perception, the current study is more representative of the postpositivist approach to social inquiry. This research attempts to explain what the subjective experiences o f community members mean to them in this context, and how meaning and the patterns of interaction that give rise to it supports or confounds the theoretical framework underpinning this study. Although subjects in this study may behave in patterned ways, th e reasons they do so are assumed to be multiple, interactive and evolving. Thus, intention and experience are unique to the individual and cannot be predicted or controlled, or generalized to other populations with any hope of accuracy. The goal here is no t to generalize, but to use multiple sources of evidence (quantitative and qualitative) to describe and explain sets of data and illuminate connections among the concepts posed, the world of these members, and society at large. In so doing, later researche rs may be able to extrapolate these findings to new circumstances, noting what is similar, what is different, and what the contextual


99 influences found here may later help explain. By providing sufficient thick description in postpositivist accounts, a rese archer offers a mechanism for transferability a means for future researchers to assess whether patterns of behavior in this context are applicable and informative of behavior in others (Lincoln & Guba, 1985). By utilizing a mixed methods approach, this re search provided at least some evidence for how people experience a particular phenomenon, and how they gather, share, and act on information about that phenomenon. This research shed light on a specific set of online circumstances involving social networks in order to advance understanding of relationship development in online groups. Research Questions Both the potentiality and experience of knee arthroplasty envelop a number of constituent elements. Some are directly related to information gathering: sol ution oriented information seeking; exchange of narratives with others (both in and outside the virtual community); analysis of pre op information; and acquisition/analysis of post op recovery information (if knee replacement is chosen); assignment of trus t and credibility to both information and source. Others encompass the more introspective aspects of the condition, including (but not limited to): realization of physical compromise and attendant identity issues; decision making; accompanying anxiety if t otal knee replacement surgery is chosen; post operation recovery and retrospection; and anxiety associated with expectations that pre condition lifestyle can be resumed. Looking to explore lived experience in this context, the following questions guide thi s research: RQ1 : How do people describe the experience of being a member of an online

PAGE 100

100 community? How does a community member define his or her role in the community? People experience the world uniquely according to his or her individual perceptions (Schu perceptions of how he or she fits into the community. RQ2 : What significance does the internet have for these community members as it relates to his or her physical (knee related) condition? Peop le are stimulated by many (or few) motivations to use the internet. As Markham (1998) observed, depending on level of commitment, people have described it as a tool a place to be, a way of life. Do these members spend time in this community to gathe r and compare only information related to knee health, or do they go there for casual interaction unrelated to knee health? Turkle (1995) noted that some people structure their lives around being a member of an online community. RQ3 : Does this online comm unity generate opinion leaders? If so, how? Opinion leadership is evident when people need information, advice, or interpretations and turn to others with whom they have some connection and feel they can trust. Opinion leaders exhibit some special expertis e or wisdom regarding the topic about which a decision must be made (Lowery & De Fleur, 1995). If opinion leaders are present in this community, how are they accorded this status in a text only environment? RQ4 : How is trust developed and maintained in thi s online community? Trust among individuals has traditionally been formed via social cues interpreted through face to face interaction (Rotter, 1967). Computerized communication media

PAGE 101

101 may be removing familiar styles of interaction, thus altering processes of trust generation. RQ5 : How is credibility accorded to source s and message s in this community? In accounts of face to face support groups, participants report that knowing others in the group have faced similar problems and circumstances lends credibili ty to advice and support offered by those members (Balk, Tyson Rawson, & Colleti Wetel, 1993; Wright, 1997; Young & Tseng, 2008). RQ6 : Is information gleaned from the community otherwise verified, or taken as fact? One may act on certain information witho ut further corroboration if it comes from to face interaction. That same information gathered in a text only setting may be interpreted differently. If so, are differences the result of the source or the medium? RQ 7 : How do community members use the health related information they gather in these online communities? There is evidence that when those managing the same chronic condition share observations with each other, their collective wisdom can yield clinical insights beyo nd the understanding of any single patient or physician ( Sarasohn Kahn, 2008). Allowing patients to tr ansport the full value of health communities back offline may hold promise for the integration of new and traditional health care services.

PAGE 102

102 CHAPTER 3 METHODOLOGY This study sought to answer the research questions utilizing a mixed methods research design. Social Network Analysis, an online survey, and online interviews are distinct methods that contribute a mix of perspectives through which the pheno mena of interest are examined. A mixed methods design is appropriate for the case study undertaken here. Mixed Methods Mixed method investigations are becoming increasingly popular in the social sciences and are considered a legitimate, stand alone researc h design (J. W. Creswell, 2008; Hanson et al., 2005; Hesse Biber & Leavy, 2003; Johnson & Onwuegbuzie, 2004). When both quantitative and qualitative data are included in a study, results may be enriched in ways that may have been constrained by using but o ne form, allowing researchers to gain a deeper understanding of the phenomenon of interest. Results of instrument based measurements may be augmented by contextual, field based information (Greene & Caracelli, 2003). Greene, Caracelli, and Graham (as cite d in Hanson et al., 2005, p. 226) identified a number of rationales for combining data collection methods. Two, however, are applicable to this research. First, quantitative and qualitative methods can be combined to recast results from one method to quest ions or results from the other (they termed this initiation); s econd, combining these approaches can extend the breadth or range of inquiry by using different methods for different inquiry components (expansion). Mertens (2003) added that mixed methods inv estigations may be used to better understand a research problem by converging numeric trends from a sample of a population and

PAGE 103

103 using them to identify individuals who may expand on the results through qualitative methods. Filstead (1979) contends the most important aspect of integrating these two forms of data is that qualitative methods provide the context of meanings in which the quantitative findings can be understood, rendering substantive significance to apparent statistical associations. Lincoln and G uba (1985) suggest there are many opportunities to utilize quantitative data in a qualitative study, but the purpose of qualitative data is to maximize information, not facilitate generalization. Lincoln and Guba also propose that the use of different meth ods of data collection offers contextual validation, thus providing varying degrees of triangulation (p. 306). Clark (1998) asserts that triangulation of quantitative and qualitative methods in the same research indicates tacit acceptance of postpositivism as researchers here recognize the diversity of truths reachable through different forms of inquiry, but valuing each for the contribution they can make to knowledge development. Case Study the term is used broadly (Babbie, 2004), this research certainly focuses attention on a limited number of instances of a social phenomenon. Yin (1994) defines a case study as an empirical enquiry that uses multiple sources of evidence to investigate a con temporary phenomenon within its real life context. Case studies have been conducted in education (Hurst, 1991), AIDS prevention (K. Fox, 1991), management science (Stake, 1995), anthropology (Geertz, 1960, 1965), and health policy (Walsh Childers, 1994). Case study researchers may seek only idiographic understanding of a particular case, or as with grounded theory, cases can form the basis for the development of more

PAGE 104

104 general nomothetic theories. This research is better described as representative of the ex tended case method (Burawoy et al., 1991) in that it attempts to modify existing social theory instead of simply approving or rejecting it. This study looked not only for observations that supported the theories presented herein, but also for observations that conflicted with these theories and offered opportunities to modify them. Knee replacement surgery (knee arthroplasty), or its potential, provides a platform for an examination into how members of a specific type of online community gather and make mea ning of information they share on this and other topics as any community might. These members include (but are not limited to) those who have experienced partial or total knee replacement and wish to share these experiences with others, those gathering in formation to supplement existing knowledge in order to arrive at an informed decision regarding the necessity of such surgery, and those researching for a friend or family member. Unobtrusive observation, an online survey, online interviews, first hand de scription of lived experience, and textual a nalysis of interview data will provide insight into this phenomenon. As the two step flow and opinion leadership are inarguable mechanisms of social structure, it makes sense to explore these concepts in terms of the relational ties and the patterns of ties that link members of a social structure in this research, a social structure in the form of a community created and maintained in virtual space. Social Network Analysis as Method As Social Network Analysis (SNA ) is a component of this research, a description of its elements and a summary of its development in the social sciences and appearance in mass communication literature is in order. SNA is both a theoretical p erspective and a set of methods (McCarty & Bern ard, 2003). In terms of theory, SNA

PAGE 105

105 begins with the assumption that members of a group interact more than would a randomly selected group of similar size, and the features of t hese relations extend and comple ment traditional social science by focusing on t he causes and consequences of relations between people and among networks of people rather than on features of individuals (McCarty & Bernard, 2003). As Burt (1980) points out, network analysis is a potentially powerful methodology for connecting micro an d macro levels of social theory. A social network can be defined as any bounded set of connected social units. This definition highlights three important characteristics of social networks. First, networks have boundaries; some criterion exists to determin e membership in the network. Second is connectedness ; to be part of a social network, each member must have either actual or potential links to at least one other member of the network. Third is social unit or unit of analysis; n etwork analysis can be easi ly applied to a wide range of social units (Streeter & Gillespie, 1992). They can be individuals, as in the case of social support networks, however, they can also be social service agencies, social institutions in local communities, or nations in the glob al economy (Scott, 2000). SNA looks beyond the specific attributes of individuals to consider relations and exchanges among social actors, inquiring about interactions that create and sustain social relationships. Actors have a set of interdependent relat ionships embedded in a social or organizational system which influence behavior and relational ties are channels for the transfer of resources Network models focusing on individuals view the network environment as shaping individual action and conceptua lize structure as lasting patterns of relations among actors (Wasserman & Faust, 1994). The types of resources exchanged among actors can be many and varied; tangibles, such as goods

PAGE 106

106 and services, or intangibles, such as influence and social support (Wellm an, 1997). In a computer mediated environment, resources are those that can be communicated to, or exchanged with, others via textual, graphical, animated, audio or video based media (Ga rton et al., 1997). SNA reflects a shift from the individualism comm on in the social sciences towards a structural analysis. This metho d suggests a revision of fundamental units of analysis, as the unit is now the relation, e.g., kinship relations among persons, communication links among officers of an organization, friend ship structure within a small group (Ga rton et al. 1997). The interesting feature of a relation is its pattern: patterns do not display age, sex, religion, income, nor attitude, although these may be common attributes of the individuals among whom the rel ation exists. Wasserman and Faust (1994) argue that network patterns can reveal whether the network structure supports or limits individual action, and thus, individual outcomes. The concept of a social network embodies a complex set of social phenomena, and studies of social networks have examined a diverse set of properties. Streeter and Gillespie (1992) however, summarized these properties and contend they can be represented by two major categories: relational properties and structural properties Relat ional properties focus on the content of relationship s between network members and on the form of these relationships. Studies of relational properties seek to understand why a network exists and to determine the functions performed by the r elations among its members. Two prominent attributes of relation al properties are transaction content, and the nature of relationships. Transaction content refers to what flows o r what is exchanged in networks: basic types of exchange content are resources,

PAGE 107

107 information, influence, and social support. The nature of relationships refers to the qualities inherent in the relationships between members of the network, noted by importance or significance of the relationship, and frequency of interaction (Streeter & Gillespie, 19 92). A number of diverse strands have shaped the development of present day SNA. The gestalt tradition in psychology, which stresses the organized patterns through which f psychotherapeuti c methods to uncover friendship choices. Using experimentation, controlled observation and questionnaires, Moreno and associates explored the ways in and, therefore, for their personal psychological development. argument that a social group exists in a space that comprises the group together with its surrounding environment, but that this environment is not external to and independent of the group. The environment that matters to the group is the perceived environment social meaning is activ ely constructed by group members on the basis of their perceptions and experiences of the contexts in which they act. Lewin proposed that the interdependence between group and environment could be explored in mathematical terms as a system of relations proved fruitful for later research. Cartwright and Harary (1956) built models illustrating the systematic interdependence between attitudes held by different individuals within a

PAGE 108

108 group. These researchers outlined the basic idea of representing groups as collections of points connected by lines; the resulting sociograms, or graphs, represent the actual interpersonal relations among group members and can be analyzed, they argued, by using mathematical ideas of graph theory. While researchers were encouraged by results from small group investigations using sociometric ideas (Roethlisberger & Dickson, 1939; Warner & Lunt, 1941), the first application of sociometrics to large scale socia l systems explored the spread of disease from one individual to another through chains of contacts and aimed at the derivation of predictive epidemiological models of contagion (Rapoport, 1952). Homans (1950 ) synthesized earlier work on SNA around the ide a that human activities result in human interaction, with interactions varying in frequency, duration and direction, and that interaction is the basis on which sentiments develop among people. Homans introduced an updated form of matrix logic that helped i llustrate the structural patterns inherent in group interaction. By separating the forms of relations from their contents, Nadel (1957) argued that the general features of structures can be described and investigated through a comparative methodology utili zing a mathematical approach. Nadel contends that social structures are structures of roles and roles are defined through networks of interdependent activities. Further deconstructing the concept of social network, Mitchell (1969) focused on individuall y anchored partial networks, identifying individuals and then tracing their direct and indirect links to others. This research concentrated on particular individuals and the content of their relations on ego centered networks focused around particular type s of relationships. Interpersonal networks, Mitchell claimed, can be at least partially

PAGE 109

109 measured through concepts of reciprocity, intensity and durability, and a social density derived from a translation of graph theory into sociological languag e describes the extent to which all possible relations in a network are actually present. Two later mathematical innovations helped firmly establish SNA as a method of structural analysis (Scott, 2000). First, development of alg ebraic models of groups led to a reconsideration of earlier work in graph theory, and role in social structure. While Nadel contended that social roles were central in social network analysis, Lorrain and Whit e added that many roles display structural equivalence general types of social relations maintained by particular categories of social actors exist. The idea behind structural equivalence is that once the uniformities of action that define social roles are identified, the networks of relations existing between these roles can be explored. The concept of structural equivalence is germane to the current study in that there may exist members in BoneSmart who, although their personal knee related experiences m ay differ, can be defined as opinion leaders, a category of social actor who interacts in particular ways with others. The second innovation was the development of multidimensional scaling, a technique for translating relationships into social distances a nd for mapping relationships in a social space utilizing relational data (Levine, 1972). Bringing SNA current, McGrath, Blythe, and Krackhardt (1996) showed how sensitive data interpretation is to the particular visual configuration presented. These resear chers concluded that if analysts wished to infer something about the sociometric properties of

PAGE 110

110 the network, then the physical distance between graph points should correspond as closely as possible to theoretical distances between these points in the graph. Online Surveys G reat advances were made during the twentieth century in the techniques and technologies utilized for survey research, from systematic sampling methods to enhanced questionnaire design and computerized data analysis (Evans & Mathur, 2005) The field of survey research has become more scientific and, over the last 25 years in particular, technology has revolutionized the way s in which surveys are administered. In the late 1980s and early 1990s, prior to widespread use of the Web, e mail was e xplored as a survey mode (Schonlau, Fricker, & Elliot, 2002). As with the Web today, e mail then offered the possibility of nearly instantaneous transmission of surveys at little or no cost. Unlike the Web, however, early e mail was essentially static, con sisting of basic text only message s that were delivered via the internet. E mail surveys tended to resemble the linear structure of a paper survey and were generall y limited in length (Evans & Mathur, 2005). Further because e mail surveys were text based document formatting was rudimentary at best. The only substantial advantage they offered over paper surveys was a potential decrease in delivery and response time and cost, although some observers also hypothesized that the novelty of the new medium migh t actually have enhanced response rates (Y. Zhang, 2000) The Web started to become widel y available in the early to mid 1990s and quickly supplanted e mail as the internet survey medium of choice (Schonlau et al., 2002). As an increasing amount of commun icative activity takes place through this new medium, there has likewise been a substantial increase in scholarly research on virtual

PAGE 111

111 communities, online relationships, and a variety of other facets of computer mediated communication ( Carter, 2004; Flahert y, Pearce, & Rubin, 1998; Preece, 1999; Preece & Ghozati, 2001 ; Walther, 1996; Walther, Gay & Hancock, 2005). In the two decades since the first electronic surveys gathered data over the internet, scholars have come to recognize both the numerous strength s and potential weaknesses of this form of research, even as the medium continues to evolve. Evans and Mathur (2005) cite several strengths of internet survey research: Global reach, flexibility, speed and timeliness, technological innovation, convenience, ease of data entry and analysis, low administration costs and ease of follow up. Another major advantage (and critical to this research) is access to unique p opulations An online survey can utilize the ability of the internet to provide access to groups and individuals who would be difficult, if not impossible, to reach through other channels (Garton et al., 1997; Wellman, 1997). In many cases, communities and groups exist on ly in cyberspace (Wright, 2005). I t would be difficult for example, to find a large, concentrated group of people conducting face to face discussions of topics such as cyber stalkin g, online stock trading, and the pros and cons of knee replacement. While people certainly discuss such issues among friends, family members, and co work ers, few meet face to face i n large groups to discuss them. Virtual communities offer a mechanism through which a researcher can gain access to people who share specific interests, attitudes, beliefs, and values regarding an issue, problem, or activity (Ta nis, 2008; Wright, 2005). Health communication researchers have studied online health related populations, including examining how features of the computer medium help peopl e cope with health condition s (S. Fox, 2005; Warner & Procaccino, 2007).

PAGE 112

112 As with a ll methods of data collection, concern s exist wi th online surveys ( Schonlau et al., 2002 ; Wright, 2005 ). Although not limited to internet surveys (yet still a concern), the validity and reliabil ity of self reported data can be issues (Bryman, 2001). A nothe r major concern with online su rveys is low response rate. In recent meta analyses, response rate for W eb based surveys was found to be approximately 34%; about 11% lower than for other survey techniques ( Shih & Fan, 2008; Singh, Taneja, & Mangalaraj, 2009) A commercial study found the median response rate to be 26 % ( Hamilton, 2009). Reasons for low response rates include : Perception as junk mail, respondent lack of online experience/expertise, technological variations (type of internet connection and conf security issues According to several meta analyses, the salience of a topic is one of the most important factors influencing response rates in both mail and W eb surveys (Fan & Yan, 2010). W hen the topic is of high interest to those exposed to the survey, potential respondents are more likely to respond. A survey is a means of gathering information it is not an end product. A survey is often conducted because it may be the only means of obt aining information (Babbie, 2004). While commercially oriented surveys are typically geared toward decision making and executive action, academic and professional disciplines most often conduct surveys to answer research questions and test hypotheses about the propensities and predispos itions of people in efforts to enhance theoretical and conceptual knowledge in a given discipline (A lreck & Settle, 2004). For the current research, a surv ey is the quickest, most efficient and economical way to understand wh at members of a unique population generally think about knee

PAGE 113

113 arthroplasty, internet use, others with whom they interact in cyberspace, and what being a member of this community means to them. This survey collected original data from a unique population too large to observe directly. Results were evaluated in light of information obtained through other methods, and treated as an additional body of evidence or set of indications. Online Interviews According to Mann and Stewart (2000), the main tools of data collection favored by those doing qualitative research are interviewing, observation and document analysis. Interviews are often used to verify, validate, or comment on information obtained from other sources (Lindlof & Taylor, 2002). While they can take a number of forms, interviews are distinguished primarily by whether they are standardized (structured), or non standardized (unstructured). Standardized interviews utilize uniform and unvarying questions and restrict answers to a limited set of response ca tegories. Although considered unstructured, non standardized interviews can employ a semi structured format in which either an interview protocol organized into specific thematic areas is utilized, or a branching method is used whereby spontaneous probing questions arise from a small initial selection of more open ended questions (Mann & Stewart, 2000). The main advantage of semi structured interviews is they offer purposive topical steering (Flick, 1998, p. 106), as this format allows interviewers to track the issues within the phenomena of interest that are of most interest to them. Also non standardized, yet in contrast to semi structured interviews, are in depth interviews, which are completely unstructured and place greater emphasis on the subjective experiences of individuals (Denzin, 1989). Evoked by a broad inquiry from the interviewer, the form and content of extensive reflective responses are structured by

PAGE 114

114 participants in the course of in depth interviews. Although the interviewer provides the foc us of the interview, prominence is given to the stories people tell, the voice in which stories are told, and narrator as knower of the self (Seidman, 1991). One goal of the qualitative interview is to obtain descriptions about the lived world of the inte rviewees with respect to interpretations of the meaning of the described phenomena (Kvale, 1996). Another is to improve understanding of social and cultural phenomena and processes rather than to produce objective facts about reality and make generalizatio ns to given populations (Fidel, 1993; Pettigrew, Fidel, & Bruce, 2001; Wang, 1999). Over the years, however, researchers have identified challenges associated w ith observation and interview methods, including cost, time, and limited access to research part icipants (Denzin & Lincoln, 2005; Kvale, 1996; Miles & Huberman, 1994; Patton, 2002; Strauss & Corbin, 1998; Taylor & Bogdan, 1998). Challenged with the ta sk of identifying new methods and tools for conducting more effective research while retaining or imp roving quality, researchers started to explore use of the internet for carrying out qualitative research Researchers began to use and still use three major internet based qualitative research interview types : online synchronous, online asynchronous (utili zed in the current study) a nd virtual focus groups. Usually conducted via e mail, online asynchronous interviews, unlike e mail surveys, are semi s tructured in nature and involve multiple e mail exchanges between the interviewer and interviewee over an ex tended period of time. This type interview is also different from virtual focus groups i n that the information volunteered by individual

PAGE 115

115 participants is not shared with, viewed, or influenced by other participants (Schneider, Kerwin, Frechtling, & Vivari, 2002). McCoyd and Kerson (2006) cite several a dvant ages for both researcher and respondent in the use of asynchronous e mail i nterviews For respondents, the format allows for completion of the interview at his or her convenience and with no visual cues to create judgment ; it occurs without the pressu res of face to face interaction. For researchers, the format offers e xtensive, longitudinal communication and provides responses already in textual form. Additionally, the drawn out nature of e mail intervi ewing (interviews typically require 8 14 interactions) allows for follow up questions and fully complete interviews in ways that the more immediate, one shot nature of most face to face and telephone intervi ews do not (McCoyd & Kerson, 2006). Althoug h e m ail interviewing limits research to those with internet access, the method, on the other hand, democratizes and internationalizes research (Meho, 2006). In contrast to face to face and telephone interviewing, e mail interviewing enables researchers to stud y individuals or groups with special characteristics or those often difficult or impossible to reach or intervie w face to face or via telephone. Asynchronous online interviews have been conducted with populations as diverse as corporate executives (Lehu, 2 004), prosthesis users (Murray, 2004; Murray & Sixsmith, 1998), self reported self injurers (Hodgson, 2004), stroke survivors (Murray & Harrison, 2004), women with breast cancer (Hybye, Johansen, & Tjrnhj Thomsen, 2005), people with disabilities (Bowker & Tuffin, 2004), knee injury sufferers (Maloney Krichmar & Preece, 2005), and members of populations who are geographically dispersed (Carter,

PAGE 116

116 2004; Foster, 1994 ; Karchmer, 2001; Olivero & Lunt, 2004) or located in dangerous or politically sensitive sites (Meho & Tibbo, 2003). Some researchers, however, have had less success with online interviews. Hodkinson (2002) reported that using semi structured asynchronous online interviews lacked the fruitful mutual interaction he experienced when conducting face t o face interviews. One result of this lack of interaction was that his e mail interviews involved fewer exchanges and participants did not become as engaged as he had hoped or expected. Bennett (as cited in Mann & Stewart, 2000, p. 76) preferred real time chat to either asynchronous e mail or face to face interviews because she valued the immediacy of real time responses not only for the speed at which her enquiries could be addressed, but also because chat allowed a negotiation of meaning between herself a nd her co researchers while avoiding potential unease trying to discuss meaning during face to face interaction. Overall e mail interviewing offers opportunity to access, in an interactive manner, emotions, and memories in their own words (Meho, 2006). It allows for the recording of anecdotes that p articipants share to enhance accounts of t heir experiences, and for participants to cons truct their own experiences through dialogue and i nteraction with the researcher. Semi stru ctured e mail interviewing is additionally empowering to participants because it allows them, in essence, to control the flow of the interview enabling them to answer at his or her convenie nce and in a manner suitable t o them (Bowker & Tuffin, 2004). Meh o (2006) advocates using a mixed m ode interviewing strategy when possible, however, e mail interviewing is a

PAGE 117

117 viable alternative to face to face and telephone interviews, especially when time, financial constraints, or geographical boundaries are barriers t o an investigation. Phenomenology A goal of this research wa s to understand lived experience. People express their experiences of phenomena through language spoken and written. Phenomenology provides for the systematic exploration and description of the e ssence and meaning of lived experience (Wolff, 1999). It looks at people's everyday experiences of phenomena and how these experiences are structured, focusing analysis on the perspective of the individual experiencing the phenomenon. Its ultimate aim is i dentification of the basic structure of a phenomenon based upon the convergence of accounts among any number of subjects, achieved through phenomenological reduction finding emergent themes in descriptions of the phenomenon (Bull, 2001; Schutz, 1967; Willi g, 2008). A central qu estion then, becomes, What characteristics are typical of this phenomenon? Textual Analysis In this study, phenomenological inquiry relies on other sociological traditions in the quest to illuminate behavior in this context. Throug h textual analysis (Ryan & Bernard, 2003), SNA contributes again as it provides the connective framework to help understand the relational aspects of this phenomenon. Textual analysis itself is an analytical tool that produces concepts, categories and prop ositions, resembling the grounded theory approach originated by Glaser and Strauss (1967). The mechanics of the process are straightforward: produce verbatim transcripts of interviews (online interview data are already in written form); read through text; identify potential themes that arise; as analytic categories emerge, pull data (exemplars) from those categories together and compare them, considering not only what text belongs in each emerging

PAGE 118

118 category but also how categories link together. Use the rela tionships among categories to build theoretical models, constantly checking the models against the data particularly against negative cases. Throughout the process, keep running notes with an eye toward emergent themes and new directions for the research (Ryan & Bernard, 2003). Some analysts suggest looking for metaphors, repetitions of words, and for shifts in content and information about how people solve problems. The object is to look for metaphors in rhetoric and deduce underlying principles that mig ht produce patterns in those metaphors. For example, in a series of studies on American marriage, (1982) method was to exploit clues in ordinary discourse for what they tell us about shared cognition to glean what people must have in mind in order to say the things they do Quinn found that the hundreds of metaphors in h er corpus of texts fit into just eight linked categories, which she labeled : lastingness, sharedness, compatibility, mutual benefit, difficulty, effort, success (or failure), an d ri sk of failure. informants often compared marriages (their own and those of others) to manufactured and durable products (it was put together pretty good) and to journeys ( we made it up as we went along; it was a sort of do it yourself project). Qui nn saw these metaphors, as wel l as references to marriage as a lifetime propositi on, as exemplars of the overall expectation of lastingness in marriage. Reliability and Validity For the traditional quantitative researcher, human phenomena are studied scien tifically by converting them into numerically measurable data. For the qualitative researcher who wants to study subjective experience, restricting data to measurable variables is unnecessarily limiting (Filstead, 1979). Qualitative research assumes the

PAGE 119

119 be people most always relate their experiences in storied form (Auerbach & Silverstein, 2003). When social scientists construct and evaluate measurements, and interpret data resulting from va rious methods of data gathering, attention must be paid to the concepts of reliability and validity. Whereas reliability means achieving consistent results from the same measure, validity refers to attaining results that accurately reflect the concept bein g measured. Reliability and validity are important criteria for evaluating quantitative research as these concepts are intended to ensure that the measurement tools and techniques utilized are objective. For the SNA matrix and questionnaire components of this study which utilize numerical data to indicate and summarize trends specific to this research established measures were utilized to help ensure reliability and validity (Table 3 1). Reliability and validity have different connotations for qualitative data. As this research employed mixed methods, one of its assumptions was based on the notion that while quantitative methods can help guide the research, they cannot fully access the phenomena in which this investigation is interested (i.e., lived experi ence, the nature of social interaction, respondent perspective). Three criterion for producing justifiable, legitimate and credible qualitative research have been offered by Rubin and Rubin (1995). 1) Transparency is achieved when research is conducted an d inscribed with enough rigor and detail that allow subsequent researchers to make accurate comparisons and determine if results are applicable in other contexts. Essentially this entails keeping a permanent record of each step in the research design and p rocess, of the original data used for analysis, and of the

PAGE 120

120 the thought processes involved and assess the accuracy of conclusions; 2) Communicability is attained when theme s and constructs in the research are recognized current study, such as opinion leadership, trust, credibility, tie strength, etc., stem from communication and sociological theories that have previously been explored in other contexts. Member checks were performed to help validate findings; 3) Coherence ensures that themes and constructs in a study are internally consistent, that they fit into an established system of beliefs relationships among statements (Rubin & Rubin, 1995). The major constructs in this study are indeed related: trust, credibility, and influence are components of opinion leadership, homophily is relat ed to effective communication, which in turn is related to opinion leadership, and centrality in social contexts implies opinion leadership. Trustworthiness is a construct offered by other scholars that overlaps Rubin and e ving viable research. Lincoln and Guba (1985) include credibility, transferability, dependability and confirmability as criterion within their definition of trustworthiness that, when applied systematically, increase the potential for producing legitimate and credible qualitative research. Definitions This study explores several concepts in an effort to understand the research phenomena. Some are not measured numerically and are therefore defined nominally, providing clarification on how these terms are use d in this research. Measurable concepts require operational definitions, specifying how they are to be measured (Table 3 1).

PAGE 121

121 Nominal Definitions VIRTUAL COMM UNITY A network of individuals who register and interact on a specific w ebsite via the internet u sing text and photos to discuss and act on mutual, interests, ideas, feelings and goals. C OMMUNITY MEMBER A person who participates in website activities as a registered member of the website. For this study, persons are registered as members of BoneSm art.org. A CTIVE MEMBERS Registered BoneSmart.org members who have posted at leas t once during the past 120 days (A BoneSmart parameter and used by many forums.) T HREADS Text messages that are grouped together v isually in a hierarchy by topic on a web site. O PINION LEADERS Defined as active BoneSmart community members who exert or have exerted influence on other members. Operational Definitions To aid in the development of specific research procedures that would result in empirical observations rep resenting measurable concepts relevant to this research, the following operational definitions are offered: I NFLUENCE A n indicator of opinion leadership; the ability of a member of the BoneSmart site to affect decisions made by other members. Influence was determined by combined SNA centrality measures (high number of direct connections as represented in matrices or graphs [Scott, 2000]), sociometric mapping (high number of nominations via Items 12 13 in questionnaire [Katz & Lazarsfeld, 1955]), and inte rview responses that specified particular members. T RUST T he belief or confidence in the honesty, integ rity, reliability, and fairness of other community members, and that the BoneSmart website operates in the best interests of community members. This was measured by summed scores on a 14 item, 5 point Likert scale (Ridings et al., 2002) and discussion with interviewees. C REDIBILITY The belief a qualifications and/or abilities to know the truth about a topi c, and a judgment about summed scores on an 8 item, 5 point semantic differential scale (Measure of Source Credibility, J. C. McCroskey & Richmond, 1996) and discussion with inter viewees.

PAGE 122

122 H OMOPHILY T he degree of perceived similarity a community member ascribes to another member occurring through textual interaction in terms of attitude, morality, and background. This item was measured by summed scores on a 9 item, 5 point semanti c differential scale ( A Four Factor measure of Perceived Homophily J. C. McCroskey et al., 1975) and discussion with interviewees. C ENTRALITY Describes the condition of members with high numbers of direct connections or ties, to other members in a con structed matrix or graph (Scott, 2000). Network analysts are often concerned with describing the "strength" of ti es. But strength can have several meanings, thus its meaning is inextricably linked to study design. One definition of strength is frequency of interaction; do actors have cont act daily, weekly, monthly, etc. Another definition is intensity, which usually reflects the degree of emotional arousal associated with the relationship. Although f requency like intensity, is a relational property in a social network, it do es not identify beha vioral ties (such as those in this study) which account for emotion. For this study, intensity was chosen to define tie strength, as this property best reflects the concept of opinion leadership in this context I ntensity provides a means to distinguish among several levels of interaction (e.g., general comments, asking for advice, experiential proximity) assigning value to ties, as opposed to simply counting the number of interactions between members. Sampling an d Recruitment While quantitative studies largely employ random sampling methods to guarantee that the sample selected is representative of the population of interest, qualitative researchers are not concerned with sampling for the purpose of generalizing t o other populati ons (Lindlof & Taylor, 2002). Instead, qualitative researchers focus on recruiting participants who are knowledgeable and experienced with phenomena under study. While this mixed method investigation utilizes both quantitative and qualitati ve data,

PAGE 123

123 qualitative data is given priority in that the experiences of respondents can be more fully accessed and understood via interviews. Looking to describe not only what is common and patterned among community members, but also what is individual and variable within this network, this study utilized purposive sampling, selecting the BoneSmart Knee Forum community due to the large number of active members frequently conversing on the topic of knee arthroplasty After approximately 18 months of periodi cally reading posts (no registration, no interaction, unknown to all except the forum administrator ) in order to determine how research worthy the BoneSmart the investigator secured permission to surve y and interview members with owners of BoneSmart.org and the administrator questionnaire occurred through a letter of introduction from the forum administrator of the BoneSmart consent form. Recruitment for online interviews occur r ed via an invitation to participate at the end of the questionnaire. The BoneS mart Knee Forums claim 2,481 members (667 active) and offer comprehensive information regarding knee pain, injury, replacement and surgery recovery, and provide a network for its members to discuss a variety of on topic issues. While this site provides We bcasts, links to reference and video libraries, and step by step preparation guides for any and all forms of knee surgery, members also interact on other subjects both in the three Knee Forums and in the BoneSmart Social Room.

PAGE 124

124 Registration is required in order to interact with other members a nd moderators on BoneSmart.org w ebsite. Splitting the difference between response rate for online surveys of 34% (Shih & Fan, 2008; Singh et al., 2009) and 26% (Hamilton, 2009) (see above), expected N for the online s urvey component of this research was 200 (30% of 667 active members). Procedures This study utilize d several methods of data collection and analysis By observing member postings in the knee forums on the BoneSmart.org website, a matrix comprised of a per centage of active members on the BoneSmart site employed Social Network Analysis ( SNA ) to determine if members were connected directly or indirectly by one or more different relationships. The online survey indicated in general and limited terms wha t memb ers of a virtual community thought about knee arthroplasty, internet use, others with whom they interact in cyberspace, and what being a member of this community meant to them Online interviews sought to capture the experience of knee arthroplasty (or its potential) and what membership in a virtual community meant to respondents. SNA Observation The BoneSmart.org w ebsite offer s pre and post operative knee and hip replacement patients two main forums one for knee, one for hip in which members can ask ques tions, give answers and share experiences These forums total 4,022 members and although participation among members between these forums is fluid, the three knee forums contain 62% of total threads. The three knee forums incorporate approximately 2,481 m embers, or about 61% of total members, with 667 active as of this w riting T here are 1,270 threads containing 15,568 posts in the Knee Replacement

PAGE 125

125 Pre Op F orum, the Knee Replacement Recovery F orum includes 3 002 threads with 46,066 po sts, and the Robotics/ Minimally Invasive Surgery F orum 23 and 346 respectively, totaling 4,295 threads (as of this writing). To determine if certain members were more central in communication networks across the knee forums via the number of ties they have in the knee forum co mmunity, a 200 x 200 adjacency matrix was constructed (Scott, 2000). An adjacency matrix lists each member twice one in the rows and once in the columns. The presence of a tie between members is represented by an X in the appropriate cell of the matrix. On the BoneSmart site, threads in each forum are numbered and listed chronologically in descending order (most recent first), and postings within threads are numbered and displayed likewise. To avoid chronological bias, a random sequence generator was used to randomize numbers representing the threads in each forum. Threads in each forum were then opened sequentially in the randomized list. T he number of postings in each thread (postings are numbered chronologically) was noted and a random sequence generato r again used to scramble numbers representing the posts. T he medi an post in that thread was copied and pasted into a separate file, and that procedure repeated until a list of 200 (30% of 667 active members) distinct screen names, associated with the orig inator of the post the source of the relational tie was generated. If an even number of posts occurred within a thread, the numeric value below where the median would have been (nearer the initial number in the list) was selected. The Knee Pre Op Forum con tained 30% of total posts across the three knee forums, with the Knee Recovery Forum holding 69% and Robotics/MIS Forum 1%. To obtain 200 screennames and accurately represent forum proportions 30% (60), 69%

PAGE 126

126 (139), and 1% (1, down from 2 to arrive at 200 ) of the screennames were gleaned from the Pre Op, Recovery, and Robotics/MIS Forums respectively. The process revealed 26 overlaps between the Pre Op and Recovery Forums, so threads and postings in those two forums were revisited using methods and percentag es previously described until 200 distinct screennames completed the matrix. T he Robotics/Minimally Invasive Forum contained only two po sts as per the sampling process and t hese posts repeated informa tion found in other Knee Forums. Since the posting membe rs were represented in one of the two other forums and these posts would have added nothing to the research they were ignored. The adjacency matrix is presented as an Excel spreadsheet: a cell containing an X indicates the presence of a relationship (a ti e); empty cells indicate no tie between those screennames in this sample (Example in Table 3 2 ). To determine ties among members within The Pre Op and Recovery Forums separately, and also distinct from the cumulative matrix, the first 60 discrete screenn ames from the Pre Op Forum, along with coded postings, were copied and pasted into an Excel file as a spreadsheet. The source and target were noted, and c ontent of the posting, indicating tie strength, was coded as follows, based on Bales on Process Analysis : 1 = general comments 2 = thanks/well wishes 3 = support 4 = seeking advice/information 5 = giving advice/information

PAGE 127

127 6 = experiential proximity 7 = giving advice/info + experiential proximity (5 + 6) That process was repeated for the first 139 discrete screennames from the Recovery Forum. a forum administrator and a n a ssi stant moderator who, in addition to interacting with members about knee health, maintain these forums with re gard to rules, protoc ol, and posted content, and serve as intermediaries between BoneSmart staff and members on technical matters. If, during the random selection process, the post of record was from the forum administrator or assistant moderator, a nd tha t post had anything to do with the business or administrative component of the forum, it was eliminated, and the thread n ext in the sequence was instead retrieved Only posts originated by the forum administrator or moderator that directly related to knee health were included in the sample). Those two Excel files were saved as text files (necessary for UCINET ), then imported separately into the NetDraw application of UCINET6 resulting in two distinct sociograms visual representations of ties among membe rs in these forums (Figure 3 1). Degree a component of centrality is a measurement that describes the location of members in terms of how close they are to the "center" of the action in a network though definitions of what it means to be at the center di ffer. If a member receives many ties, they have high in degree and are said to be prominent or to have high prestige That is, many others seek to direct ties to them, and this may indicate their importance. Members who have unusually high out degree are those who are able to exchange

PAGE 128

128 with many others, or make many others aware of their views. Members who display high out degree centrality are said to be influential members. B asic meas ures of centrality were performed on the sociograms generated from k nee forums, with results indicating certain members having high degree, or being well connected. Online Questionnaire A 54 item online questionnaire was disseminated in the co mmunity after the a researcher The quest ionnaire was accessible to community members for a period of three weeks (In a 2009 meta analysis, Hamilton found that for online interviews, 96.5% of responses arrive within two weeks of introduction.) The questionnaire component for thi s study was desig ned and implemented via VT Survey, version 2.2.4. VT Survey is hosted on a secure server provided by the College. On the Knee Forums page of BoneSmart website, clicking on a one radio button then directed respondents to the questionnaire. In addition to demographic information (Items 44 55), the questionnaire include d measures of general internet use (Items 1 4, Papacharissi & Rubin, 2000), community involveme nt (Items 7 11), knee health related information gathering (Items 5 6), trust (Items 14 27, Ridings et al., 2002), credibility ( Items 28 34, J. C. McCroskey & Richmond, 1996), and homophily ( Items 35 43, J. C. McCroskey, Richmond, & Daly, 1975). Seeking op inion leaders, the questionnaire also ask ed respondents to name (via screen names) who in the community they s eek out for general information (Item 13) and who they go to in the community for information about their own particular circumstance ( Item 12, f ollowing

PAGE 129

129 the sociometric method of determining opinion leaders introduced by Katz & Lazarsfeld, 1955). E mail addresses were solicited for follow up interviews. Online Interviews Important in this research is the range of variation in the attributes of co mmunity members. Questionnaire results guide d the selection of 25 ( J. O. Brown et al., 2007; Davis, 2006 ; Morse, 1994 ) follow up respondents who volunteered for interviews and who represent ed the widest possible range of circumstances in the community. (J. O. Brown et al. [2007] utilized purposive sampling to achieve the widest range of demographics possible for their study, ultimately interviewing 30 respondents Davis [2006] weighted her selection criteria to reduce interviews from 85 to 21 Both these st udies utilized SNA and reported the number of respondents recruited was adequate to generate ample data. A single case may be sufficient to display something of substantive importance, however, Morse [1994] suggests us ing at least six participants for stud ies in which one is trying to understand the essence of experience. Morse also suggests 30 50 interviews for ethnographies and grounded theory studies .) Selection of 25 respondents from among those who completed the questionnaire and volunteered for interv iews was governed by variation in the distribution of individual attributes produced by the questionnaire. The investigator divulge d more specific information about the nature of the research to these respondents and conducted semi structured asynchronous e mail interviews. This format allowed p articipants to enhance accounts of t heir experiences and to cons truct their own experiences through dialogue and interaction with the researcher. Semi structured e mail interviewing also allowed participants to contr ol the flow of the interview to some degree. Online interviews offered opportunity to unders

PAGE 130

130 past ex periences as they related to knee related health condition, obtain descriptions of events normally unavailable for observat ion, and to understand the circumstances under which influence occurs. Research Questions Revisited While e mail interviews with volunteer respondents provided analytical depth to the research questions posed in this study, some dimensions of particular questions were answered in part by way of responses to questionnaire items and the SNA matrix. In this study, Research Questions 1, 6, and 7 were answered through semi structured, open ended questions whereby respondents construct narratives of their experiences as they relate to the q uestion. Research Questions 2 5 were guided (and partially answered) either by SNA measures or res ponses to questionnaire items. See Table 3 1 fo r delineation of major concepts This study utilize d sev eral methods of data collection in efforts to answer seven research questions Observation of member postings in BoneSmart.org knee forums allowed Social Network Analysis ( SNA ) to determine if members were connected by one or more different ties and to asc ertain if certain individuals were central to communication pro cesses in the community. An online questionnaire was designed and implemented to establish wha t members of the community thought about forum leadership, knee arthroplasty, internet use, and to capture demographic information consisting of questions addressing trust, opinion lea dership, credibility, and use of information, were conducted via e mail. These interviews a lso included a phenomenological component, seeking to understand the experiences of community

PAGE 131

131 interview responses. Results from these procedures in this mixed method appr oach shed light not only but also revealed of leadership in an online environment and delineated the physical and emotional experiences common to those suffering this condition. Results from eac h procedure are detailed in the Findings chapter, which follows.

PAGE 132

132 Table 3 1. Operational definitions of major concepts answered by concept operations SNA survey items interviews reference influence SNA c entrality measures Scott, 2000 sociometric mapping X 12, 13 X Katz & Lazarsfeld, 1955 trust summed scores on a 14 item 14 27 X 5 point Likert scale Ridings et al., 2002 credibility summed scores on a 8 item 5 point semantic differential scale 28 34 X J. C. McCroskey & Richmond, 1996 h omophily summed scores on a 9 item 5 point semantic differential scale 35 43 X J. C. McCroskey, Richmond, & Daly, 1975 centrality high number of direct connecti ons with other members in matrix X Scott, 2000 Table 3 2. Example of SNA matrix S1 O2 M3 R4 M5 S6 N7 R8 S1 O2 M3 X X R4 M5 S6 X N7 R8 X Note : Rows represent the source of the relationship tie columns represent target

PAGE 133

133 Figure 3 1. Sociogram of Knee Pre Op Forum (partial ) Note: ine thickness and color aid tie type recognition. D irection of arrowhead indi cates direction of relation Location of node is arbitrary

PAGE 134

134 CHAPTER 4 FINDINGS Th is study investigated how the concepts of opinion leadership, trust, and credibility operate in an online community centered on knee arthroplasty and how people here describ e the experience of being a member of this community. Social Network Analysis (SNA) provided a method to observe communication patterns in the community, while an online questionnaire and e mail interviews offered opportunities to ceptions of interaction in the community. General findings are reported first, then detailed under each data collection method, followed by more specific accountings in answer to the research questions. Through observation and SNA analysis, certain communi ty members emerged as opinion leaders. This finding was confirmed not only by responses from members who participated in the online questionnaire, but also from responses to semi structured questions posed to questionnaire respondents who agreed to be inte rviewed via e mail. T rust, credibil ity, and homophily are elements generated by individual perception that help promote voluntary online cooperation among strangers in virtual communities. While scores measuring these elements were relatively high across the board, members still claimed to verify through alternative sources the knee related information they accumulated from community interaction, although experiential proximity (shared knee conditions or circumstances) was the most noteworthy generator of trust and credibility. These findings also suggest that an informed patient makes for a better patient. Virtually all BoneSmart members who participated in the e mail component of this study said that by gathering from others in the community pertinent i nformation relating

PAGE 135

135 to his or her personal situation helped them ask more focused and relevant questions of health care professionals, which led to more prepared and ultimately healthier decision making. SNA Observation In an effort to initially determine if certain members of this virtual community were central in the interactions among members which may indicate opinion leadership observational techniques common to SNA were employed. An adjacency matrix was constructed to determine simply who among memb ers in this sample had ties. The existence of a relation between two members is considered independently of whether or not the relation involves influence one way or the other. The data consists of binary (presence/absence) undirected data. Certain members (via screen names) were more prominent in the 200 x 200 adjacency matrix via degree greater numbers of associations with other members indicating centrality and possible leadership. Simply counting the number of in ties and out ties of members suggests th at certain members are more central, and it appears that this network as a whole may have a few central members. In undirected data, as in this adjacency matrix, members differ from one another only in how many connections they have. This matrix revealed P RBX76 (32 ties), PRBN66 (19 ties), PRFZ182 (8 ties), PRCS97 (5 ties), and PRAE31 (5 ties) to be most prominent in terms of degree. These five accounted for 9.8%, 5.8%, 2.4%, 1.5%, and 1.5% of total posts this matrix represents. With directed data that is valued, however, it can be important to distinguish centrality based on in degree from centrality based on out degree. Generally, if an actor in a network receives many ties, they have high in degree (summed incoming ties based on tie strength compared to others ) and are often said to be prominent ; many

PAGE 136

136 other actors seek to direct ties to them, and this may indicate their importa nce (Hanneman & Riddle, 2005). In this research, members with high in degree are either those who are receiving advice about their condition or circumstance in response to a query, have recently undergone a procedure related to knee health and are receiving various levels of well wishes and support, or are opinion leaders on a knee related topic and are being queried. Members who hav e high out degree (summed outgoing ties based on tie strength) are those who exchange with many others and make others aware of their views. Members who display high out degree centrality are said to be influential actors, and for this research they can be opinion leaders depending on the strength of the ties. Using the NetDraw function within UCINET6 valued data fro m both the Pre Op and Recovery F orum Excel files were used to generate sociograms that illustrated the direction and type of tie for each fo rum. These visualizations enable an investigator to analyze social structure; opinion leaders in a social system can be identified in both matrices and maps, depending on the nature of the investigation. Directed ties among sampled members of the Pre Op F o rum are shown in Figure 4 1 Figure 4 1 illustrates social structure in this sample of the Knee Pre Op F orum. Summing out going ties by tie strength for each member shows that PRBX76 (orange), PRBN66 (green), and PRCS97 (yellow) are clearly central in this sample, signified by high out degree centrality. Note that while PBW75 (gray) and PRAE31 (red) have fewer ties than the three members mentioned above, they also demonstrate high out degree based on stronger tie strength stronger value of their outgoing adv ice. PRBX76 exhibits both high out degree and high in degree centrality (summed incoming ties by

PAGE 137

137 tie strength), indicating an influential member in this community. Note that while PER148 (black) has high in degree based on the strength of the advice he/she is receiving, this member displays no out degree. Running Freeman raw displays centrality measures for the Knee Pre Op Forum ( Table 4 1 ). To help understand these data, it is useful to consider the standardized scores for i n and out degree measures (Freeman standard deviations the individual score is above or below the mean for these data. The normalized scores for PRBX76 further confirms centr scores are more than 6 and 2 standard deviations above the mean for in and out degree respectively. Figure 4 2 illustrates social structure in this sample of the Recovery F orum. Again, PRBX76 and PRBN66 are central, al ong with, to lesser degrees, RCU99 (yellow), RFG163 (aqua) and PRFZ182 (red). Also note that while RBD56 (gray) has high in degree based on the strength of the advice he/she is receiving, this member displays no out degree ties. As the Recovery F orum is larg er than the Pre Op F orum, measures of degree centrality are proportionately greater for those considered central. See Table 4 2 for Recovery Forum degree centrality measures. It should be noted that PRBX76 is the one paid forum administrator (though n ot the only moderator) for the BoneSmart Forums; a circumstance that may affect the frequency and strength of messages directed to and emanating from her simply by virtue of position. As previously mentioned however, only posts originated by the forum

PAGE 138

138 adm inistrator that directly related to knee health were included in the sample. Any post that had to do with business or forum administration was eliminated, and the post next in the sequence was instead retrieved. This research focu sed only on communications form ing the core of this investigation: the exchange of various types of information directly related to knee replacement and recovery. While PRBX76 demonstrated both high in and out degree centrality, others in this research who also demonstrated central ity were not moderators ( as of this writing ) BoneSmart conducts a vetting process t o qualify those who assume forum administration and moderator positions. In the flow of information, the structure of a social network helps determine the network's usef ulness to its individuals. Observation of this social network suggests that the attributes of individuals may be less important than their relationships and ties with other members within the network. Online Questionnaire An online questionnaire was dissem inated in the forum to d etermine how these members gather knee health information, why they interact in this particu lar forum, and if, and to what degree, trust, credibility, and homophily are factors here. In addition to demogra phic data, information pe rt inent to this study also included measures of time spent online how participants spent that time, how long participants had been members of the BoneSmart community and why they joined, who they chose as leaders in the community, and how similar to themse lves they feel others here are. Descriptive characteristics of survey respondents revealed that o ve rall, respondents were primarily older, college educated Caucasian women who were small town U.S. residents, employed full time in professional or technica l occupations, and members of the BoneSmart com munity for three months or less (Table 4 3). Several

PAGE 139

139 respondents reported, however, that they lived in other English speaking countries. Results showed that trust among members is fairly high and that the info rmation exchanged in the community is believed by these participants to be credible. In addition, in item on the questionnaire. Fifty two members of BoneSma to the 54 item online questionnaire. One member failed to answer over half of the 54 items and was th erefore discounted in analysis Several questions addressed internet usage using a 5 point semantic different internet 5 = This is what I use the internet for). Mean hours spent online per week for any reason other than work was 24.08 weekly mean for BoneSmart hours 4.53, and mean hours per week looking for knee related health information not in BoneSmart was 2.23. Highest mean score for why these participants used the internet 4.27 and 4.10 respectively, wi th using the internet of how they gathered knee health related information, the BoneSmart w ebsite was most used at 2.0 (rank ordered; 1 most used, 10 least used scores averaged ) with Reasons for joining the BoneSmart post point

PAGE 140

140 ommunity they perceived themselves to be, mean score was 3.58. Critical to this research, however, were nominations of certain community community. Given the opportunity to l ist four distinct leaders a mai n source knee contact/leader then three additional leaders distinct from other nominations depicted in Figures 4 1 and 4 2 Table 4 4 displays findings for sociometric data. Since this study suggests trust, credibility, and homophily are elements of individual perception ( J. C. McCroskey & Richmond, 1996) that help promote voluntary online cooperation among strangers in virtual commun ities (Tanis, 2008), results from measures of these components in this context are central to its findings. Fourteen trust related questions (based on Ridings et al., 2002) were measured on a 5 point Likert scale (1 = Strongly Disagree, 5 = Strongly Agree ) and assessed specific trust information, concern for other members, and ability to know the topic under discussion. ust topics in general outside of the community. Mean for all trust responses was 4.02. As this study is concerned with how opinion leadership and influence are assessed through communication in a text only environment, and trust is a component of effective communication, it is helpful to understand how trust is built over time in this community.

PAGE 141

141 Table 4 5 displays trust item means for survey participants based on length of time as a BoneSmart member, and shows a pattern that develops over time. Except for a decrease in seven of the ten measures in the median membership category in Table 4 5 trust generally increases across all measures the longer a participant remains in the community. The trust measure that fluctuates and yet shows a decrease from short term to long term members are looking for information, and long termers have been through knee replacement and remain in the community to furnish information. Likely due to a small N in each group, differences were not statistically significant ( oneway ANOVA; F[4, 44] = 1.08, p=.377) and trust could not be predicted by length of membership (linear regression ; F[1, 47] = 3.28, p=.076). Credibility was assessed by seven questions using a 5 poi nt semantic differential scale, e.g., U ( Table 4 6 ) Previous research shows that the relationship between the receiver and the source affects the degree of credibility accorded to the source (Aune & Kikuchi, 1993). Me ans for credibili ty measures here were fairly high (4.13) supporting the notion that credibility in this community largely derives from the perception that the source of information has been through similar circumstances, has had similar problems and enga ged in similar behaviors, and has similar attitudes and beliefs about the con dition he or she is facing. When people share common meanings, beliefs and mutual understandings, communication among them is more likely to be effective, and they are more likely to develop and maintain supportive relationships with others to whom they are similar on important social dimensions (Lazarsfeld & Merton, 1954). Homophily is the degree of

PAGE 142

142 perceived similarity a receiver ascribes to a message source. Homophily was assess ed with nine questions based on bipolar scales developed by J. C. McCroskey et al. (1975) ( race, religion, age and sex may account for changing perceptions of homophily and h e terogeneity, education is also a n important variable influenc ing homophily (Marsden, 1988). As this questionnaire did not assess religion, and less variation exists in dimensions of race, age, and sex than in education, this study assesses homophily by co mparing means of this measure according to educational subgroups. Most participants perceived that they were communicating with people like themselves, as evidenced by the relatively high mean scores for these measures ( Table 4 7). The largest perceived d ifference in similarity was recorded by the participant with the highest reported level of education for the variables ng m ay not be meaningful Results from the online questionnaire suggest that people in this forum do recognize that certain members among them are leaders in the ways they present themselves online. These findings also show that members perceive others with whom they communicate h ere to be similar to them in many ways and that the information offered is credible and trustworthy. Online Interviews At the conclusion of the 54 item online questionnaire, participants were presented with an opportunity to take part in online interviews by offering a valid e mail address through which interviews could be conducted. In an effort to capture the widest possible range of demographic characteristics, dispositions, and knee health related

PAGE 143

143 experiences, as well as narrow the interviewee list to 25, careful consideration was given to the thirty four survey participants who provided e mail addresses (one proved invalid). The final 25 selected for interviews included, for example, a high income university professor with a doctorate and low homophily scores who had simultaneous double total knee replacement, and a lower income skilled laborer in the community less than three months looking for information for an upcomi ng left total knee replacement. By means of e mail exchange in response to semi str uctured questions, interviews were conducted to not only understand the experience of potential and performed knee arthroplasty, but also to confirm or deny the existence of opinion leaders in this community in Moreover, interviews were structured to extract whether or not these concepts are present here and if so, how they develop and are maintained. Textual analysis revealed underlying themes associated with the physical aspects of this phenomenon, and direct responses to questions indicated that trust and credibility are developed and maintained through thematic confirmed that opinion leaders in fact do exist in this community and are almost exclusively the same ones who became apparent both through SNA observation and nomination in the questionnaire. Introductory e mails were sent to those 25 (plus an additional four to allow for attrition) explaining the interview process, acknowledging their agreement to take part in the interview process, and asking them to reply if still interested, with 17 responding. So as not to intimidate interviewees with an overwhelming amount of information at one

PAGE 144

144 time, interview questions were sent in three rounds, each two days after rece ipt of responses from the prior round Interview questions were designed in part to help answer the Research Questions posed in this study. Round One consisted of four multi part questions addressing onli ne trust, opinion leadership, and credibility of information; with three multi part questions, Round Two focused on what being a member of this online community was like and how individuals used the information they gathered; Round Three (condition questio ns) consisted of several questions seeking to understand how the physical changes and accompanying anxiety affected the il this written reflection ( Appendix C contains full list of questi ons). Of the 17 responders, 10 completed all three rounds of questions, one completed two rounds, three completed one round, and three answered no questions. Those who did not complete all three rounds were sent two follow up reminders, which went unans wered. Textual Analysis Participants were coded (e.g., N10, G7) and responses from each participant for each question were aggregated into one of seven separate word files, corresponding to which Research Question the response addressed. In the process of identifying expressions that seemed important and then arranging these into blocks of related text, thematic categories emerged. Repetition is one of the easiest ways to identify themes; some of the most obvious themes in a body of data are those topics that occur and reoccur. The more the same concept occurs in a text, the more likely it is a theme (Ryan & Bernard, 2003). Others themes are more abstract and require the researcher to synthesize characteristics of participant experiences according to the

PAGE 145

145 theoretical understanding of the phenomenon. To understand lived experience through the perspective of the individual experiencing this phenomenon, responses to condition questions were analyzed apart from those designed to answer specific Res earch Questions. Several themes arising from textual analysis were practical and more obvious than others. One was issues associated with body weight. While apparent in our society in general, decreased mobility for several of these participants was strong ly related to weight gain, especially among older women; two participants stated that they were obese. Another was assignation of leadership to a recurring set of individuals, confirmin g results from both SNA observation and questionnaire results ( Figures 4 1 and 4 2, Table 4 4 ). A third practical theme linked active participation in the BoneSmart community with becoming a better, more informed patient. Certain themes, however, were subtle and required somewhat more abstract thinking to discern linkages among expressions. When scripting about courses of action undertaken in efforts to verify information and sources, an encompassing theme surfaced indicating that most participants put more stock in the real life experiences of peers with regard to pain ma nagement both prior to and following surgery, and in peer reported recovery therapies, than they did in scientific information on those subjects, even ahead of advice from health care professionals in most cases. Another less concrete theme that emerged un derscored the concept of influe nce. While all answered negatively t o the question of whether they were directly influenced by someone in BoneSmart to have (or not have) knee surgery, a recurring theme in the data was how members read, filtered, then follo wed the advice of other members on a variety of knee

PAGE 146

146 independently. Research Questions Resolved This research offered some evidence for how people experience a specif ic phenomenon, and how they gather, share, and act on information regarding that phenomenon. Using research questions designed to explore these experiences, this study expanded on a particular set of circumstances involving online networks in order to adva nce understanding of relationship development in cyberspace. For each Research Question, selected quotes from respondents that illuminate the concept of inquiry appear, followed by a synthesis of themes extracted from all responses to that question. Remain ing quotes are located in Appendix D. RQ1 : How do people describe the experience of being a member of an online community? How does a community member define his or her role in the community? of how he or she fits into the community (Schutz, 1967). In general, most participants went to BoneSmart for information about how other members dealt with like issues. Many claimed that being a member had been a large part of their lives in recent month s or years, and for those now on the other side as members refer to their post surgical life, they still participate so they can be with like minded individuals with whom they share a common experience. Most logged on daily, some out of habit, some out of a genuine desire to connect with other members who helped see them through a tough time in their lives. P4 I

PAGE 147

147 question this fall about what an x ray showed. I do learn things, still. I go to see if there are new interesting subjects or new parts to the story of one of the members. When I first joined BS, I was desperate and worried and needed to know if anyone had ideas about what was wrong with my leg. I needed to know what was normal and what meant something was wrong with the knee. I needed them to help me. D13 : During my recovery per iod I was on the forum several times a day, in fact, it was the first site I visited when I woke up. I needed the support I received from the members. now and then but the majority of my time spent there was to overcome the isolation I felt. M9 : Being a member of an online forum is a great experience. I meet people all over the world who share very similar problems as I do. Since it is if I want to stay in my pajamas, I can. BoneSmart fits into my life easily, since I can go online ded family and look forward to checking on many Most noteworthy in response to the first part of this question were the way s in which members described their initial need for help and how they then fo und the help they sought through others in similar circumstances. Other themes included the satisfaction some felt in giving back to the community by offering their post surgical perspective, and how being a member of this online community was an easy fit into their lives as they compared their recovery to similar others and kept track of fellow That members fall along a op and seeking information) to senior (usually past single or double knee replacement) means that a wide array of experiences can be articulated in the BoneSmart forums, affecting how others experience the forum. On the topic of how a participant felt he or she fit into the community, some described their role in terms of what they got out of being a community member, others

PAGE 148

148 depicted themselves as lurkers, others explained their roles through expressions of T11 : With Bonesmart, the time for needing help has passed, and I'm happy to try to help the others. It's a very sociable place, with a spine of frequent members, and this gives [ me ] a chance to develop social bonds with a lot of people. M9 : change that and start giving more. V3 : I go online to BoneSmart to read the latest threads in the knee recovery section almost every day. I go there only for knee related information I guess I would des cribe myself as a lurker, mostly reading or taking information but occasionally giving some input. through act s of giving back. Members described how they enjoyed giving advic e and help to new members and how through giving back they gre w to care about others and did what they could to help others do well. These members derived a sense of satisfaction from their role in the community. input, yet benefitting from membership nonetheless. A peripheral of RQ 1 asked members what they perceived BoneSmart to be: Is it a community ? ? N10 : ace maybe, but just a group of like minded individuals who have some common experiences with knee replacemejt [ sic ]. I think of a community as a face to face location. G7 : I refer to it [ BoneSmart ] strictly as a message board. I have a disabled spouse and I can't change up too many words or use something like "community" to explain what I'm doing. There are plenty of people who do treat it as a community. V3 thought was no, BoneSmart is no t a community. I would agree that if one

PAGE 149

149 P4 : This is a community. We give each other support and sympathy, since we totally understand what we ar e all going through. well and be able to continue with their lives in a comfortable fashion. And mfy, cozy place where I always feel welcome. A majority of members queried for this research experience the BoneSmart Knee Forum as community. These members feel that this online forum not only shares a common interest, but members ca re about each othe health progress and about each as a community. This view was not overwhelming however. Others felt BoneSmart was and what they perce ived to be the transient nature of membership member comings and goings prohibited BoneSmart from being a community. RQ2 : What significance does the internet have for these community members as it relates to his or her physical (knee related) condition? There is nothing new in lo oking for health information on line People have going to the w eb for health inform ation as long ago as the 1980s well befor e search engines made it accessible to a wider audience (J. R. Schwartz, 2008). These ex panded capabiliti es have allowed internet users to share information easily, inverting the top down flow of information between doctors and patients. Depending on level of comm itment, people have described the internet as a tool, a place to be, a way of life (Markham, 1998 ). T11 : I use the internet to get info, but apart from Bonesmart, I only use it as a tool to find something, like info or to buy something, or to solve a problem It's essentially a tool to me, although I'm happy to devote some time to being sociable

PAGE 150

150 D1 3 : The Internet was virtually my sole source for knee related information. My consultant asked me to go away for 6 months and lose some weight. I was very disheartened that I would have to spend a further 6 months in a great deal of pain and limitation. I published this online and was he then rescinded] and there was no consultation about my expectations, possible out comes, possible complications, s o my only source for any information was the Internet hence I Internet life V3 : I consider the Internet as a remarkable tool The Internet is vital to me in getting up to date information dealing wit h the ongoing problems I have with my knee. Most people may visit a site before and directly after their surgery, but when things work out wel l they move on with their lives. The small percentage of people that have problems continue to participate in the forum and at first glance this might suggest a higher percentage have problems than actually do The Administrator cautions people that the number of problems reported is disproportionate to the number of knee replacements done Thus I us e the Internet for checking this against outside stats Members were split when describing the importance of the internet in their lives. Just over one half characterized using the internet as a way of life for them, that t he internet played a major role as they suffered k nee conditio ns and online r esearch was key for them The way of life theme seemed particularly fitting for many members, as they were physically compromised to some degree and were not easily able to get to the library or other places where they might gather information rel ated to their condition Yet m any also reported that the internet was essentially one tool among others s office were also thought to be tools. One mem ber mentioned she used the internet as a tool to meet her husband on an internet dating site. Use of the internet as a tool seemed to be a more practical application of technology as opposed to the immersion of self into an ubiquitous medium. RQ3 : Does t his online community generate opinion leaders? If so, how?

PAGE 151

151 Opinion leadership is perceptible when people need information, advice, or interpretations and turn to others whom they can trust. Opinion leaders demonstrate experti se or wisdom regarding the topi c for which information is sought ( Katz & Lazarsfeld, 1955). If opinion leaders are present in this communi ty, how does leadership develop? The SNA observation and online questionnaire components of this research (above) helped answer this in part, and res ponses to online interview questions completed the picture. B12 : BoneSmart opinion leaders are gifted with the ability to use words in such a way as to build relationships and gain the trust of members. These leaders have also experienced similar experien Experience is a great relationship builder. They are genuinely concerned for the member...that they have the best possible experience in their journey. PREQ 147 sees the replacement process from a He can relate well to those (men & women) who have been athletic all their lives and are facing replacement He can by example help these folks to see that at the end of light. One can return to a normal lifestyle. He is very intelligent and by his participation over the years has gained much knowledge in the physiologic al aspects of the replacement. In real life he is a referee for football and baseball as well as a teach er RGG189 sees the replacement from experience. She never offers medical advice, b ut offers experiential advice. Yet her manner is that of a W8 : I consider PRBX76 to be a leader in the BoneSmart group beca use of her expertise as a nurse. According to these members, opinion leaders in this forum emerge primarily through experience Members feel that certain others contribute valuable inform ation not previously known, information that stems from prior resea rch and knee trauma experience. That t he information resonates with their personal circumstances lends credibility to both source and message. T he ability to communicate with sympathy compassion, and understanding, combined with a willingness to recommend that people become their own advocates and not fear

PAGE 152

152 distrusting the medical establishment, also contribute to the perception of opinion leadership. In a few cases, bona fide medical experience and credentials create the perception of opinion leadership as well. In this online community however, these qualities are always coupled with the experience of knee replacement itself. RQ4: How is trust developed and maintained in this online community? Before computers, trust among individuals was formed when peop le interacted face to face (it still happens this way). Computers seem to be changing that though. The aggregate mean for responses to the 14 trust related items in the online questionnaire was relati vely high at 4.02, however this research question attemp ted to address how trust develops. For members, how does trust in others develop in a text only environment? Why would a member trust another member when they may never meet in person? V3 oday is generally the same in a few days or weeks, I begin to place more weight on the f you get a big enough sample of contributors you are bound to get someone who has experienced similar problems (or successes) to what you are experiencing. W8 : I know that there are a number of people who post on BoneSmart whose information I do not trust. I think that the trust is developed and maintained within the BoneSmart community through the longevity of some posters. Others post over a significant amount of time, and, to me, that improves their trustworthiness U14 leading [ sic ] information by users. I believe if I were to throw out a mis leading response to someone on Bone Smart a bunch of folks would set me straight. The overriding theme emerging from responses to this question was consistency not only in textual content, but also of demeanor. Members felt trust is

PAGE 153

153 built over time through constant and reliab le behavior in textual interaction. Many respondents stated that others must demonstrate empathy and some degree of intelligence in their posts, and s eem steady, honest, supportive and caring befo re trust in either that source or his/her message is assigne d. Members also appeared to agree that experiential proximity is a generator of trust, because the experience of knee debilitation, surgery, replacement and recovery is so intense that people are not inclined to lie about it and imposters can be sniffed o ut rather easily. RQ5 : How is credibility accorded to sources and messages in this community? Participants in face to face support groups report that knowing others in the group contend with similar problems and circumstances lends credibility to advice a nd support offered by those members (Young & Tseng, 2008). While aggregate credibility and h omophily mean scores were somewhat high (4.02 and 3.45 respectively), d o members perceive information coming through a computer from an amorphous source to be less trustworthy? If so, is it less trustworthy because it comes from someone (the source) or is it less trustworthy D13 : It was only after exploring the whole site further and reading the vast in there Also the fact that the site was not asking for money or subscriptions gave me confidence in the site. I explored various Orthopaedic sites and forums but felt comfortable with the people on Bone Smart G7 : I take any information I get from an online community with a grain of salt. If I know someone else is having the same issue I am, that brings a little pe ace of mind. W8 : I do take with a grain of salt much of what I read on discussion boards like BoneSmart. Anecdotal information from other knee replacement patients is good in so far as it can alert the reader to what to expect after surgery or offer the r eader tips on how to functio n once back home from

PAGE 154

154 surgery. person. Verification was a constant theme in member responses to this question. Virtually every member stated that they checked other sources of information outside BoneSmart (websites, people who had experience knee replacement) before making decisions with regard to their knee health. If information gathered in BoneSmart matched that obtained from other sources, information was the n perceived to be credible. Some members stated that they do not take information available on the internet at face value, and although BoneSmart was their first choice for information, they look ed elsewhere for verification before assigning credibility to the BoneSmart information on which they based decisions. Experiential proximity also factored into the perception of source/message credibility. Several members revealed that if they believed other s with whom they were interacting had experienced what th ey were going through, that was enough for them and that particular other and his/ her message were deemed credible without further verification. RQ 6 : Is information gleaned from the community otherwise verified, or taken as fact? One may act on certai n i nformation without corroboration if it comes through face to face interaction from a known and trusted source. Although i nformation shared on the BoneSmart w ebsite is reportedly reviewed and authenticated by an administrator for accuracy and appropriatene ss virtually all respondents claim ed to verify information through other sources. When prompted to rank order sources from which respondents gathered knee related health information, most used to least, 38% of respondents selected BoneSmart as their firs t choice, followed by 32% choosing medical

PAGE 155

155 professionals first and 19% opting for other internet sources first. Just 4% chose friends first, while 8% chose family. Accessibility and availability may be factors here though, since doctors are not as availabl e as are the BoneSmart Forums. W8 : Once I knew that I was going to have my knees replaced, I read a consider information from the internet to be less trustworthy than print sources. W ebsites sponsored by major hospitals, for instance, seem to be pretty reliable. Having a bit of an in with healthcare professionals, I could talk to nurses, nurse practitioners etc. My orthopedic surgeon was also a source of information for me I took inf ormation I read on BoneSmart to be trustworthy and believable when I could verify it Some things did not have to be verified (i.e., how to prepare the house for recovery, what sort of clothes would be easiest and most comfort able in rehab). Trusting those postings was a matter of common sense E17 : I always check out any information that comes from the internet There is a lot of false information running around out there! Verification ag ain was the dominant theme and members here mentioned how quickly one could confirm what they found in BoneSmart via virtually instant access to information that would have taken months to locate prior to the internet Responses also indicated that the v erification process sometimes included face to face disc ussions with other knee rep lacement patients and consultations with doctors and health care professionals with regard to the surgery itself whereas p ain management and recovery therapies were more likely to be confirmed through internet sources. RQ 7: How do community members use the health related information they gather in these online communities? Given prior research demonstrating that individuals consult multiple sources when seeking health information (Rains, 2007), it appears that a lack of trust in traditional sources may be one factor motivating individuals to use online forums for medical information. E vidence exists indicating t hat when those managing a similar chronic

PAGE 156

156 condition sh are experiences with one another their colle ctive wisdom yield s i nsights beyond the understanding of any single patient or physician ( Sarasohn Kahn, 2008). While BoneSmart members cl aimed to verify information with outside sources, they also held that their decisions regarding knee health treatments were corroborated b y information exchanged in the community. D13 : Smart what had happened and the response was unanimous, why wait 6 months? The operation is inevitable so wh y suffer for a further 6 months I t was suggested that I co ntact him and ask him to reconsider and to go say what I had wanted to say all along I would say the information I have receiv ed has given me more courage to not always accept every thing as absolute just because it comes from a healthcare advisor, I feel more confident to ask questions if I have even a limited understanding of what is happening. F5 : I used much of the BoneSmart information to relieve pain and manage my post surgery reco very. Without BoneSmart I would not have known to use ice and elevation to reduce swelling and pain these measures are not advocated in [ ] I t has made me a much more knowledgeable patient. The majority of responses here integrated into Respondents reported that BoneSmart information helped make them aware of product manufacturers types of replace ment materials and procedures, what the hospital experience would entail recovery treatment, and what they could expect in terms of outcome. F amiliar ity with those issues reduced some of the anxiety and stress when consulting with doctor s and recovery professionals Many mentioned that empowerment is a direct result of BoneSmart membership they have become a more v ocal advocate of their own concerns when dealing with doctors and other health care professionals.

PAGE 157

157 Another prevalent theme with respect to the use of BoneSmart information was sharing. Some said they shared with those they knew to be in similar circumsta nces, and that they sometimes sh ared pharmaceutical advice with people they knew to be in need of that information but who did not necessarily suf fer a knee condition. Others revealed that they used BoneSmart insight to open lines of discussion with famil y members so they could be come involved in the decision making process. Revisiting the SNA observation component of this study (Figures 4 1, 4 2), and both questionnaire results (Tables 4 1, 4 2, 4 4) and participant interviews (primarily responses to Res earch Question 3), it is obvious that PRBX76 fits the definition of role in the community, an e mail interview was conducted and her comments in response to a separate se t of questions are distilled here (Full interview in Appendix D). Learning that she may need knee replacement PRBX76 joined BoneSmart in early 2007 seeking to learn about recovery ; as a nurse she was already knowledgeable about the procedure itself. Havi ng spent her entire adult life in the medical profession, she was immediately concerned at the poor informa tion members were exchanging on the site. After providing credentials to BoneSmart she began to address medical inaccuracies, and became a paid mod erator in June of 2007 then forum administrator in November 2008 People value her opinions not only for their medical value, but also because she h as experienced a r ight knee arthroscopy a l eft knee arthroscopic menisectomy and a total right knee replac ement. PRBX76 enjoys passing on her knowledge and skills to others, and provides anecdotal evidence as to why she thinks people consider her an op inion leader in both

PAGE 158

158 her online and off line worlds. She says satisfaction is her reward and feels this valid ates her sense of self worth. She claims h er leadership position provides an o utlet for her natural teaching skills as she helps people face their fears, make good choices and benefit from a less stressful recovery o n their journey to PR BX76 is convinced the BoneSmart Knee Forum is a community. By all accounts PRBX76 fits the traditional definition of an opinion leader She is a leader in this online environment, and to some unknown degree, in her offline world as well (according to he r self report). PRBX76 does disseminate information and the knowledge she has accumulated through both personal experience and from mass media (journals, c onferences, books) to other community members R ather than a two step flow of communication, however, a more accurate model may be a multi step flow. PRBX76 gathers information and directs it toward one or more individuals through a me dium that is both interpersonal and mass. While information originating from PRBX76 is deemed credible by other members, i t is frequently verified through other sources (according to these members ) before it is acted upon. Through observation of forum postings, an online questionn aire, and e mail interviews, this study found evidence for the existence of o pinion leaders tru st, and credibility in an onl ine environment, an d found that th e traditional two step flow of communication m ay now well be a multi step flow. The traditional view holds that individuals sharing geographical space engage in primarily face to face communica tion discussing topics at hand Opinion leaders are exposed to mass media messages and pass on information to opinion followers. interaction is no longer dominated by physical proximity, strong interpersonal ties, and

PAGE 159

159 face to face conversa tion, but rather seems increasingly subject to geography spanning digita l relations that involve any number of loose ties evident in this research. The use of an internet source, as evidenced here, can be seen as a further muta tion of the two ste p flow hypothesis. T he internet appears to be the first choice to inquire about knee arthroplasty and serves as a conduit for discussing those topics with other internet users Also revealed in this investigation were elements common to those suffering from knee conditions that ultimately brought them into the community: pain as an initial driver of community participation ; the impetus prompted by community interaction, to solicit second opinions reg arding his or her condition; and a wi l lingness to trust others based on perceived experiential similarities. The implications of these findings will be discussed in Chapter 5. Figure 4 1. Sociogram of Pre Op Forum members Note: is indicated numerically. L ine thickness and color aid tie type recognition. D irection of arrowhead indicates direction of relation. Location of node is arbitrary.

PAGE 160

160 Table 4 1 Op Forum (partial) outd egree ind egree nrmOutd eg nrmInd eg PRBX76 34.000 13.000 6.227 2.381 PRBN66 11.000 0.000 2.015 0.000 PRCS97 18.000 0.000 3.297 0.000 PBW75 10.000 0.000 1.832 0.000 PRAE31 8.000 0.000 1.465 0.000 PER148 0.000 12.000 0.000 2.198

PAGE 161

161 Figure 4 2. Soci ogram of Recovery Forum members Note: ine thickness and color aid tie type recognition. D irection of arrowhead indicates direction of relation. Location of node is arb itrary.

PAGE 162

162 Table 4 outd egree ind egree rmOutd eg nrmInd eg PRBX76 68.000 10.000 6.110 0.898 PRBN66 44.000 8.000 3.953 0.719 RCU99 17.000 0.000 1.527 0.000 RFG163 14.000 0.000 1.258 0.000 PRFZ182 11.000 9.000 0.988 0.809 RBD56 0.000 13.000 0.000 1.168 Table 4 3. Characteristics of survey participants Sex % Age % Ethnicity % m ale 23 18 29 0 Asian 2 f emale 67 30 41 8 Caucasian Hispanic 2 n o answer 10 42 50 25 Caucasian non Hispanic 71 51 57 19 Prefer not to answer 2 58 64 19 65+ 21 No answer 8 Residency U.S. 75% Other 25% Employed % Occupation % Edu c (last level completed) % Full time Executive, admin Secondary (high school) 23 (40+ hrs w k) 38 managerial 12 University (college) 44 Part time 15 Professional, Advn deg ( masters) 15 Not employed 15 t echnical 35 A dvn deg (Ph.D ) 2 Retired 17 Admin s upport 4 Prefer no ans 2 Prefer no ans 6 Sales, m arketing 2 Other 10 No answer 8 Skille d l abor 10 No answer 4 Other 15 No answer 23 Municipality size (pop) % Membership length (BoneSmart ) % 150,000+ 29 3 months or less 31 50,000 150,000 23 more than 3 months but less than 6 months 19 Less than 50,000 42 more than 6 months but less than 1 year 19 No answer 6 more than 1 year but less than 2 years 13 2 years or more 13

PAGE 163

163 Table 4 4. Nominations for leaders in BoneSmart Knee Forum M AIN KNEE CONTACT T ot al nominations for main knee contact % PRBX76 46.1 PRBN66 5.7 PFH164 1.9 No one 9.6 No Answer 34 LEADERS 1 T ot al nominations for leaders 1 % PRBX76 53.8 PRBN66 19.2 PFH164 1.9 RFL168 1.9 PGJ192 1.9 My physicians 1.9 No Answer 17.3 LEADERS 2 T ot al nominations for leaders 2 % PRBN66 53.8 PRBX76 17.3 PFH164 1.9 RGG189 1.9 PGJ 192 1.9 No a nswer 21.2 LEADERS 3 T ot al nominations for leaders 3 % RGG189 5.8 PRS19 3.8 PR219 3.8 PFH164 3.8 PGJ192 1.9 No a nswer 65.4

PAGE 164

164 Table 4 5. Trust by length of membe rship 3 mo or less >3 mo < 6mo > 6 mo < 1 yr > 1 yr < 2 yrs 2 yrs or more total m ean N m ean N m ean N m ean N m ean N m ean N __ get along concerned 3.92 12 4.11 9 3.88 8 4.17 6 4.57 7 4.10 42 4.15 13 4.67 9 4.25 8 4.17 6 4.43 7 4.33 43 help others 4.31 13 4.78 9 4.63 8 4.17 6 4. 43 7 4.47 43 responsive 4.08 13 4.56 9 4.38 8 4.17 6 4.57 7 4.33 43 many responses 3.23 13 3.89 9 3.50 8 4.50 6 4.29 7 3.77 43 quick responses 3.58 12 4.11 9 3.75 8 4.17 6 4.29 7 3.93 42 output priv ate info 3.36 14 4.40 10 4.40 10 3.57 7 3.29 7 3.81 48 get info on topic 4.60 15 4.90 10 3.90 10 4.43 7 3.57 7 4.35 49 give info 3.07 15 3.00 10 3.00 10 4.00 7 4.29 7 3.35 49 share skills 2.93 14 2.50 10 2.80 10 3.57 7 3.71 7 3.02 48 general faith 4.29 14 4.10 10 4.30 10 4.57 7 4.71 7 4.35 48 general reliable 4.07 14 3.90 10 4.20 10 4.14 7 4.71 7 4.17 48 general trust 4.29 14 3.90 10 4.30 10 4.57 7 4.71 7 4.31 48 Mean of means 3.78 4.06 3.95 4.14 4.20 4.02 48 Table 4 6. Descriptive statistics for credibility m easures N min max m ean stn dev reliable 45 1 5 4.04 1.086 informed 45 1 5 3.93 .963 expert 44 1 5 3.48 1.045 honest 45 1 5 4.31 .996 friendly 44 1 5 4.48 .976 pleasant 44 1 5 4.45 .975 selfish 44 1 5 4.27 1.020 valid N (listwise) 41 4.13

PAGE 165

165 Table 4 7. Means table: Homophily by education Education Homophily measures Think like Are like Personalities Share my Treat people Class Culture Ecostat Bg me me like me beliefs like I do No level given (N1) Mean 5.00 5.00 5.00 5.00 5.00 4.00 4.00 4.00 4.00 Advn deg(Ph.D)(N1) Mean 4.00 4.00 4.00 2.00 2.00 1.00 1.00 1.00 3.00 Advn deg(Mast) (N6) Mean 3.83 3.83 3.33 3.67 3.83 3.50 3.50 3.00 3.50 Std D .753 .753 .816 8.16 .983 .8 37 .837 .894 .837 Univ (college) (N22) Mean 3.64 3.50 2.95 3.41 4.00 3.10 3.41 3.05 2.76 Std D 1.09 1.10 .950 1.00 .926 1.02 .959 .999 .889 Junior college (N3) Mean 3.67 3.33 3.33 3.33 4.00 3.33 3.33 3.00 3.33 Std D .557 .557 .557 .557 .000 .557 .557 1.00 .557 Secondary (hs) (N12 ) Mean 3.58 3.58 3.33 3.91 4.27 4.00 3.73 3.58 3.00 Std D 1.31 1.08 .985 .944 .905 1.00 1.00 1.08 1.27 Total (N45) Mean 3.69 3.60 3.20 3.57 4.02 3.38 3.45 3.16 3.00 Std D 1.06 1.00 .944 .974 .927 1.05 .975 1.04 .988

PAGE 166

166 CHAPTER 5 SUMMARY, CONCLUSIONS DISCUSSION, AND RE COMMENDATIONS In part, this research reaffirmed findings from previous studies of online interaction. This study however, unobtrusively observed online behavior and then solicited reports of behavior and lived experience, which resulted in findings that ultimately amplify, expand, yet occasio nally conflict with previous research. The study revealed a variety of member perceptions in the exploration of results oriented online behavior. This chapter will examine the theoretical and methodological implications of the study where possible, point o areas of future research suggested by the findings. Summary Computer med iated communication i s not just a tool; it is a technology and a facilitator of social relations. The internet and world wide web have changed th e nature of information seeking; t he internet has become fundamental t o living in a modern day world. It has changed the way people inform, amuse, educate and care for themselves, and the ways they stay in touch with others (Pew Internet & American Life Project, 2008). I nformation consumption has become a collective process and people are using these and other innovative technologie s to pool knowledge and skills. When forms of collective intelligence develop through social interaction via the internet lo osely knit groups are formed and are referred to as online communities. Others prefer the term virtual communit y, depending on level of commitment (Rheingold, 1993). Critics argue that these types of groups lack conventional commitment and responsibility t ies, therefore intimacy and connection are inadequate for community, especially when participants can disengage when they choose (S. G Jones, 1995).

PAGE 167

167 Others contend however, that engaging and cooperating online in shared practices and information exchange enhances community In this research, shared practice manifested itself through experiential proximity mutual knee problems and knee health related conditions. Experiential proximity here fostered credibility, trust, mutual attachment, concern and in man y cases, belonging. As trust and credibility are crucial to both opinion leadership and virtual community, these concepts were investigated as components of this study. The virtual community has become a place where people acti vely seek social relationshi ps. Information that was once obtained secondhand from friends and acquai ntances (i.e., the traditional two step flow hypothesis), is now ea sily found via the internet. This study found support for the contention that the two step flow is mutating into a m ulti step flow of communication ( Case et al. 2004; Weimann, 1994) Differences between online and offline communication are important because allocating trust and credibility must be made based on written communication Past research has examined tru st an d credibility directed at w ebsites or online groups (J. O. Brown et al., 2007), however, fewer studies have been conducted looking at how these concepts develop when the intended object is perceived to be another person in a health related context. As a c omponent of the two step flow, opinion leadership has primarily been researched within the context of traditional forms of community and social interaction. People may be influenced by o pinion leaders, but the internet may be changing how opinion leadersh ip information The frequent conclusion of past research, that most people turn to others

PAGE 168

168 (family members, friends, neighbors and nearby experts, such as physicians) for advice, may be in doubt. This research sought to understand how opinion leadership, trust, and credibility operate in an online community specifically, one devoted to health related issues. It explored how these notions develop and are maintained in a v irtual environment, how me mbers he community, the internet and world wide web to help them understand their condition. Moreover, this study was designed to extract discussion s of how being a member of a particular virtual community BoneSmart.org, for med around potential, pending, and perform ed total knee replacement shapes the condition, and how they reflect on the condition itself. This topic was chosen for three primary reasons: 61% of all U.S. adults have searched for health information online (Pew Internet & American Life Project 2008 ); knee arthrop lasty is a major investment, which calls f or research on the part of the patient; and knee replacement is not life threatening, which creates conditions whereby those who join an online community may not feel as press ured to remain as they might if suffering a life threatening condition. BoneSmart.org was chosen as the research site because it is an orthopedic implant and joint re placement consumer awareness web site established in 2002 with almost 2,500 members in its three Knee Forums. The primary goal of the BoneSmart Community may need joint replacement surgery and those who have had one or more of their own

PAGE 169

169 Fund amentally a case study this research used mixed methods to ans wer seven research questions After collecting selected postings on the BoneSmart.org w ebsite Social Network Analysis (Wellman, 198 3; Scott, 2000) demonstrated that certain individuals were ce ntral to communication pro cesses in the community. This finding was confirmed not only by sociograms generated through observation, but also by responses from members who participated in the online questionnaire, and from answers to semi structured questio ns posed to questionnaire respondents who agreed to be interviewed via e mail The online quest ionnaire also included measures of trust, credibilit y, and homophily. R esults from t he questionnaire revealed an array of knee health experiences related to d emo graphic variables, which aided in the selection of 25 interview ees who represented the widest possible range of conditions and circumstance s from the sample of respondents. A na lysis of online interviews with consenting respondents th en revealed the individ ual logics these respondents employed in their com munication processes within the community. Overall, survey results indicated that members sense other community members to be trustworthy, credible, and similar to themselves, although mean scores measuri ng lower than means scores for other trust and credibility measures. While scores measuring these elements were generally high, members still claimed to verify knee related infor mation through outside sources, although experiential proximity was the most noteworthy generator of trust and credibility. Survey results also showed these participants to be more interested in gathering knee related health information from BoneSmart and other online sources than from traditional media.

PAGE 170

170 Q uestions for online interviews were disseminated in three rounds: 1) four multi part questions addressing online trust, opinion lea dership and credibility of information; 2) three multi pa rt questions fo cusing on what being a member of this online community was like and how individuals used the information they gathered; 3) several questions designed to elicit responses regarding how participants dealt with the physical changes and the anxiety as they sou ght to solve their knee problems. Interview responses confirmed and expanded upon survey results regarding opinion leadership, trust, credibility and homophily, and it appears that a lack of trust in tradit ional sources may be motivating people to seek m edical information online On this last point, interview replies also suggest that community members do not always trust a opinion after posting their original diagnosis on the BoneSmart website. Interviewed BoneSmart members also propose that informed patients make better patients, that gathering community information relating to his or her personal situation from others in the community aids them in asking more releva nt questions of health care professionals. By using the internet as a research tool, t his investigation contributes to communication research and expands the case study method The study design observation, results of which were supported by survey find ings then corroborated by direct contact with community members offers fresh evidence for how people experience a distinct phenomenon, how they gathe r, share, and act on related information, and how conventions such as leadership, t rust and credibility dev elop in a text only environment. This study also comments on the notion of

PAGE 171

171 Conclusions In a new media environment, the ways in which people form social networks, disseminate and gather information and exert influence have undergone recent and rapid change. C hanges in the communication environment have impo rtant implications for the concept of opinion leadership. People are continually creating and discovering new p laces to discuss countless topics on the internet, and discussions are no t limit ed to geographicall y proximate or even socially similar participants. This is a depart ure from the original conception in which an opinion leader exert s in fluence through and within his or her face to face social networks. The notion of centrality still ap plies, however the internet i s now established as both a primary sou rce of information and a social environment, thus the concept of opinion leaders hip is applicable to online activity. Findings, particularly through interviews, suggest that an association exists between online opinion leadership and the ability to effectively communicate leading one to conclude that opinion leadership in this environment involves the perception of possessing convincing knowledge, the ability t o influence others and a ten dency to be empathetic to others. While these traits are also present in traditional opinion leaders, what is different here is that the perception of these traits occurs in a text only environment where traditional social cues are absent. T his study also found support for theory. Questionnaire returns indicated these community members turned to family and friends far less than they did to BoneSmart and the intern et to gather information about knee related proble ms. While the internet was used in more general fashion, it appeared that the ties community members formed with other members contributed substantial ly to knee related information gain. This was confirmed through interviews,

PAGE 172

172 with emotional support and enc ouragement reported to be an additional feature of these ties (although these features are also present in family/friends ties). Several conclusions can be drawn from recognition of experiential proximity a ubiquitous concept in this research. While trus t, credibility and homophily certainly occur (or fail to occur) in offline face to face interaction among people who share a health (or any other) condition, these factors are also present in virtual communities and are fostered by experiential proximity. BoneSmart members are willing to exchange information and advice, and interaction in this community is only marginally based on like mind att itude, and background; most important is experience, as that experience relates to knee health concerns. Some fe a tures of traditional community are attained, including relationships, common purpose, s ense of identity and belonging, various norms along with unwritten rules and the potential fo r rejection. This virtual community offers possibilities for motivated and interactive communication based on experiential proximity that are not avail immediate physical environment. Experiential proximity also appears to be a factor in why people join online forums, why they remain, and is an impor world. Another important conclusion to be drawn from this research center s on the concept of community. Participants here feel that membership in the community improves their liv es in a number of positive ways. The community provides practical information to help them deal with a debilitating knee condition, complex surgery options, selection of a potential surgeon, and long periods of therapy and recovery. Membership is also a source of comfort, support, encoura gement, and various forms of

PAGE 173

173 gratitude. Even those involved in the community for but a short time report their experience as constructive and positive. E xperiential proximity is an important component in the creation of this online community. The diversity of information and experience found in BoneSmart about injur ies, treatments, recovery, and specific conditions within the fairly narrow domain of knee related health are primary attraction s to the community, and as interaction occurs on these topics it e xpands into social and personal matters, creating conditions for communal interaction on other levels. Also important is that members know they can enter the community and, through an extensive library of multi media platformed information related to knee health, take advantage of regularly occurring updates about surgery, precautions, rehab, and problems along with the occasional related video and summaries of scientific and research articles I nformation provided in the online library suppleme nts that w hich comes from fello w community members and offers a n avenue for verifying information gleaned from all sources. Knowing that credibl e and trustworthy resources are incorporated into Community is also en hanced through continuity and stability Members see others come and go; some stay for sh ort periods, some persist and become omegas (the most senior rank and file members via postings), some return periodically to share overcoming their infirmity. The w ebsite is always there 24 hours per day, seven days per week, and members ta ke comfort in knowing that the community is too ; survey participants indicated they received quick responses to their posts Lastly, empowerment is a theme that emerged in this r esearch. Interacting with others in the community appears to give m embers a sense of empowerment that helps

PAGE 174

174 th em communicate more competently with their healthcare professionals They feel they can go to their physician or specialist fortified with informa tion about many aspects of the condition they suffer, from diagnosis through therapy and pain management and discuss them intelligently. Through community interaction t hey are also able to assess t he quality of care they receive compared to other members and, in some cases seek second opinions based on that interaction. Discussion When comparing results o f this study to those of previous research in critical areas, both common and conflicting findings emerge. Studies covering online health information se eking, the strength of weak ties, virtual community, and opinion leadership were examined, as were those investigating social network analysis, the two step flow, trust, credibility, and experiential proximity. Certain findings from these studies are of co nsequence to this research. S. Fox ( 2008 ) found that reliance on internet search engines such as Google have become the de facto second opinion for patients seeking health information online particularly after a diagnosis This study supported that find ing, as BoneSmart community members reported that they searched online for information pertaining to their condition, but once they joined the community, they learn ed as much or more from each other as they did from sites offering information alone. These wide ranging capabilities allow people to easily share information, which in effect challenges the traditional top down flow between doctors and patients This permutation can be seen in part as a democratizat ion of health care. In a later study, S. Fox ( 2010 ) also found that it is not just convenience that draws health information seekers to the internet but the positive experiences they have with

PAGE 175

175 online research and interaction. This research found similar results in that t he kinds of health informatio n sought and found online (in this case knee health) ar e different from what people garner from most traditional sources, especially in the realm of personal experience, and this translates into much more than doctor patient communication. The internet and community that it can engender helps health seekers tap into resources wherever and whenever they need them connecting with whatever source they believe will help them at any given moment. This research demonstrates that t he source is not always a doctor or even a health professional. Rains (2007) reported that a growing distrust in health care provision has increased the potential for using the internet related specifics This study echoes that finding, and goes further in suggesting that a primary reason for people joining health related support communities is to gather diverse opinions about his or her condition, and they are especially cognizant of opinions that contradict their particular diagnosis or the prevailing medical position on their particular problem. Several interviewees articulated a distrust for the medical establishment. labor market study found that when information is unavailable through close, personal relationships, or strong tie s, people m ay obtain it through weak ties relationships characterized by infrequent contact, lack of emotional closeness, and no history of reciprocity. to face contact, this research found that weak ties cul tivated online can offer people access to resources not found in their strong tie relationships, since the potential for Burt ( 1983 ) su ggests that weak ties tap div erse groups, which make s for a diversity of information,

PAGE 176

176 which in turn makes those ties useful. While this study found support for that in theory, practically it was the diversity of physical conditions within a narrowly defined g roup an online group with knee problems that mad e those weak ties useful. theory operates in conjunction with the concept of disintermediation ( A. L. Shapiro & R. C. Shapiro, 1999). Disintermediation is generally associate d with the ability to engage in direct commerce without brokers, distributors, and retailers, but the concept can be expanded to describe the way technology allows individuals to bypass resource gatekeepers and redirect power from established sources. Blog s, online forums, internet gambling, e commerce sites, Usenet groups and the tools of social media are but a few modes of technology t he internet and world wide web have spawned that allow for information decentralization Users of these platforms are abl e to bypass traditional sources of information on virtually any topic singularly, or through webs of relationships they build with dispersed sources. The theory of disintermediation is relevant to this research in that participants of online health forums such as BoneSmart use these peer to peer networks to gather and act on information outside the purview of the medical establishment. Ridings et al. (2002) defined v irtual co mmunity as a collection of individuals with common interests and practices who comm unicate regularly in an organized way via the internet through a common location or mechanism While this is an apt definition for the community under investigation in this study, more important is the relationship the individual has with the activity or subject in question. This research found, similar to 2001 ), that t

PAGE 177

177 psychological self self image, thus the more likely the individual is to pursue and value membership in the community. Physical mobility is a vital component of identity for participants in this study, and frequent visits to the BoneSmart community in the pursuit of solutions to their knee problems, or to help others with their quests, were important to these discuss ed re. In so stating, Carter suggested tha t they are actively producing a social space making it a cultural rather than technological construct, mediated through experience rather than through tech nology. Participants in this research also emphasized the social aspects of this community in these ways. Carter, however, argued that hers was community simply because her participants said it was. The current study could not fully support that finding, as a less than unanimous 70% here described BoneSmart as a community (and for various reasons). vitriolic in their denunciation of random members who they felt detracted enough from the spirit of community that it could not reasonably be referred to as one. Katz a nd Lazarsfeld (19 55) defined opinion leaders likely to influence other definition remains more or less unchanged in studies of the two step flow of communications. Numerous studies sugges t that opinion leaders and followers alike are exposed to a mix of interpersonal and media influence (Watts & Dodds, 2007; Weiman n et al. 2007) Current findings indicate that participants here, while reporting that

PAGE 178

178 legitimate opinion leaders do exist in this community, also utilized media sources as well as interpersonal sources (technologically mediated and face to face) in attempting to investigations of opinion leadership and t he two step flow, is that influence, in some shape or form, is the essential concept in the theory, yet many methodologies are employed to discern its presence, with the internet adding an additional layer to the mix. The major difference between conventio nal and network data is that conventional data focuses on actors and attributes; data generated for the SNA component of this study is network data which focus es on community members and the relations among them. A dimension dissimilar to that in traditi onal assessments of opinion leaders hip is that in this online environment those who may not have been recognized as opinion leaders in the past may be here due to their experience s having equal chan ce for expression regardless of demographic attributes th at may serve as hindrances to information exchange in face to face interaction. Traditionally, p hysical appearance or perceived social status in face to face interactions often function as that control human interaction. Members in this study repea tedly mentioned that experience was the a ttribute most sought after and the one that had the most potential to influence. While t his study did use sociometric methods in part to identify opinion leaders, unlike traditional methods opinion leaders were al so located by pa tterns of relations among members in the system and not by demographic attributes. Also unlike traditional methods, the type of relationship was valued which provid ed a means to quant ify via value both incoming and outgoing messages. Opini on leaders were identified by high in and out degree scores.

PAGE 179

179 With influence still the focus, J.O. Brown et al. (2007) examined whether websites themselves acted a s social proxies for opinion leadership In that study, participants commonly mentioned that website content ( textual content ) projected a similarity of interests to them, detached from any perceived human affiliation. In contrast, members of BoneSmart deemphasized other forms of content (e.g., the online library, sponsor mentions) while accentu ating member interaction. A powerful notion in social science is the idea that individuals are bound up in webs of relations and interactions (Borgatti et at., 2002). S ocial N etwork A nalysis (SNA) has the potential to provide explanations for numerous soc ial phenomena, and studies reviewed for this paper in part reveal how autonomous individuals create social order within social space. Looking at a social network formed in an online community devoted to sports, M.G. McGrath (2009 ) found that members there, like those in the current study, no longer preferred to act as passive receivers of information transmitted through traditional communications channels. Instead, they increasingly relied on online networks of friends, colleagues, like minded individuals, and in this online sports community, other members. Moreover, these members, again like those in the current study, view these networks as more trustworthy and impartial than corporate and traditional medi a sources The theoretical link between these stud ies is the notion that alert organizations BoneSmart along with its corporate sponsors) recognize that it may prove fruit less to compete with the strength of peer networks and must ins t ead work with them. generated content by facilitating several

PAGE 180

180 user forums (the Knee Forum among them), and also sponsors clinical and scholarly research that can be accessed by members. In an SNA study of farming practices in India, D. K. Kim et al. (2007 ) established that while mass media sources of information demonstrating innovative farming techniques were readily available, farmers were influenced more by their personal contacts with local opinion l eaders, although mass media functioned as a source of new knowledge Similar f indings emerged in the current study, in that BoneSmart members rely more on experiences of their fellow members than on published medical accounts or on a dvice from medical professionals in some cases, even as they scour the internet for information related to their condition. Although D. K. Kim et al. (2007) analyzed social networks of Indian farmers, the study also revealed obvious elements of the two st ep flow of communication (Lazarsfeld et al., 1948). The current study suggests a multi step flow may more aptly describe the that those who used the internet for health information rated inform ation they obtained from family and friends to be significantly less useful than informa tion obtained from the internet. Rains also reports that thos e who did not use the internet rated information they obtained from family and fri ends to be more useful. W hile non internet users were not solicited for the current investigation, findings did reveal that family and friends were much less useful for knee related health information than were internet sources for these participants. opinions were highly regarded in this research, virtually all participants in this study reiterated their need to verify information gathered through the community. As put forth earlier in a modern world the two step flow of

PAGE 181

181 communication may be in doubt The original two step flow hypothesis (Lazarsfeld et al., 1948) ideas often flow from radio and print to the opinion leaders and flow from them to (p. 151). Numerous examples of a multi step flow surfac ed in this research however, most not ably in responses to interview questions. Several participants ( F5, M9, V3, D13, H15, W8, G7 ) alluded to a flow of information that originated with offline medica l professionals, continued through that offered by BoneS mart opinion leaders, and ebbed as it was filtered through the The result of this flow was a synthesis of information leading to confid ent and informed decision making (Figure 5 1). Adhering to the concept of a multi step flow, other communication patterns were also observed. While members are certainly not aware of how all information flows among BoneSmart interpersonal sources, research shows that potentially influential communication did flow from media sources to an opinion leader ( who infused those ideas with his or her knowled ge and experience) and then into the BoneSmart co mmunity, which users then verified both within BoneSmart and through other sources (Figure 5 1). Trust and credibility are obvious facilitators of effective communication. Much literature on trust asserts that trusting beliefs in face to face circumstances may for m quickly, be fore social actors have any sort of meaningf ul information about each other, through perceptions of s ocial categorization, disposition, or out of an immediate need to cooperate o n an undertaking (McKnight et al., 1998). In contrast, BoneSmart members repo rt that since immediacy is generally not a factor in the search for solutions

PAGE 182

182 to knee problems, and interaction is text based, trust develops over time through consistency in postings and replies. Naquin and Paulson ( 2003 ) found that a shared se nse of iden tity among actors also promo te s trust through deeper understanding of Findings in this study echoed that, and like Lewicki and Bunker ( 1995 ), this study found that shared identity essentially yiel ds more trusting relationships In an early study of computer mediated communication, Walther (1995) found that impressions of the communication partners As the number of messages multiply, people learn more about their online pa rtners, and the a dditional information help s people circumvent the lack of nonverbal cues. As a result, Walther contends, relationships resemble face to face interaction, providing a mechanism for assessing credibility. The current study supported that fin ding, the association again evident in BoneSmart reporting that c onsistency in replies and postings enhanced perceptions of the credibility of the online othe r. This finding is important in that it illustrates the differences in how trust and cre dibility develop and are maintained in an environment where social cues that would be indicators of trust and credibility in face to face circumstances (voice inflection and tone, physical gestures, body language, eye contact, dress) are absent. Tanis (2 008) found that credibility is also generated in part when a source of information appears embedded in a group and dedica ted to its success Many virtual communities emphasize this characteristic (BoneSmart included) by providing supplementary information about members, such as the length of memb ership (i.e., ) o

PAGE 183

183 attributes which distinguish an individual along a time and/or participation continuum (Huffaker, 2009). This finding is theoretical ly logical yet in the current study, while community members perceived other members to be quite credible overall, this measure fluctuated inexplicably over the five length of membership categories. Perceptions of the credibility of other me mbers was high for the newest members, deviated among lower scores in the middle three categories, then curiously rose to its uppermost mean score for the longest tenured members. No comprehensible pattern emerged, differences were not statistically signif icant, and a confounding variable might be that there were twice as many respondents in the new member category than in any other. Galegher et al. (1998) argue that a component of credibility is authority. These researchers claim that social actors somet imes referred to medical or scientific research in their online messages as a means to establish some degree of authority thereby enhancing credibility Like Galegher et al., the current research found that although reference to medical research is judged appropriate and relevant, i t is not necessarily sufficient for establishing credibility. Informat ion based on personal experience was perceived to be mor e relevant to community members, but was not sufficient to establish a uthority in and of itself BoneS mart members verified virtually all information through other sources, and were in agreement that medical advice from other members is not credible. (A few members with medical backgrounds are present in the Knee Forum, and while the medical and scientifi c information they occasionally post may be credible, they urge others to further verify that information.)

PAGE 184

184 In the final analysis, the most important idea to emerge from this study, a concept that may well drive future social science research in a world i ncreasingly reliant on technologically mediated social relations, is the notion of experiential similarity, or experiential proximity. This concept stems from the theory that similarity initiates connection. Finding others with whom one shares a similarity can be important. There are times when stru ctural similarity the demographic attributes of age, sex, race, ethnicity, education, socioeconomic status, and geography often assumed to be important in explaining interpersonal relations are of less importa nce than experien tial proximity for individuals experiencing life altering circumstances. When searching for similar others the online environment, it may well make it easier for people to concentrate on the issue in question when these attributes are not discernable, as they would be in face to face interaction. The combination of experiential proximity and a growing pool of those who might be subject to it has consequences that might explain why people choose to make use of health related online forums and join virtual communities dedicated to particular health conditions. At least some users of online forums are experiential experts, which makes them credible sources of information for the topic at hand. Forum participants find or ask for information an d/ or advice, and learn from the fi rst hand experiences of others. Information gathered here is useful base of knowledge, which in turn provides opportunity for more control and helps reduce uncertainty, which then help s people cope with their personal circumstances. Forum users find information about v arious types of s urgery and treatment, address reports coming from established medical sources, and glean information about how to cope with the uniqueness of his or

PAGE 185

185 her own condition In m any cases, restricted mobility leads to an appreciation of the opportunity to extend networks house. When people become ill, injured or infirmed, they tend to rely on health information provided by health care providers, espec ially physicians. People in these situations, however, also desire interaction and connection with others who share their experiences (Tanis, 2008). Under these circumstances, experiential similarity may trump structural similarity, as people often want to share information with experiential peers. This is certainly the case in the BoneSmart Knee community. People with knee problems can be said to occupy isomorphic positions in society, and if they connect, as many do in online health forums, contexts in w hich homophilous relations can form are created in social space. Experiential similarity is thus a component of homophily. Online communities can provide members with support and advice from a virtual forum of experiential peers, unrestricted by physical distance or time constraints t hat may, in many cases, replace that of health care professio nals. Hence, online inf ormation exchange among the experientially similar may be motivated by a combination of need, convenience, shared relevance and mutual unders tanding. Online health forums also provide emotional support and offer users the opportunity to share their stor y with others. This affective form of suppor t is characterized by displays of compassion, reassurance, and encouragement or at least by express ions of s ympathy and unde rstanding for what the other has gone through or may be going through Social companionship is also evident to some degree, in that interactions occur not directly related to knee difficulties but that intentionally distract from those difficulties In onlin e forums, there are often feeling s expressed

PAGE 186

186 through textual communication that suggest community ( Preece & Ghozati, 2001; Wright & Bell, 2003 ), and those feelings are present in the BoneSmart forum, but the sense of commu nity is not unanimous. Those with knee p roblems face challenges beyond the various medical concerns. Dealing with pain and discomfort are primary, but many report fluctuating emotions, altered relationships, shifting personal roles and identities, and dis turbing limitations in lifestyle due to decreased capabilities. They must also learn to manage the condition and its treatment in the context of their lives Information gathered through experiential proximity helps people navigate aspects of the condition beyond the scope of medicine, such as choosing a surgeon, managing his or her own treatment and dealing with problematic social situations brought on by the condition. Physicians can provide facts, but other patients can relate what it really feels lik e and what to expect next, in way s that only someone with personal experience can (Preece, 1999). Much of the information BoneSmart members glean from the community is not that which one can learn from health professionals, rather it is from sufferers like themselves who possess this form of knowledge. BoneSmart members share personal stories about their respective experiences and, in part, this involves making medical information meaningful to others by linking medical concepts, such as prescribed m edicat ion side effects, to personal experience. In the recovery phase, much discussion revolves around medication regimens and the success of one's treatment or side effects. T he narrative detail they apply to treatment and recovery issues makes complex informa tion more intelligible to others. This speak s to the value of peer based information exchange. M oreover, when strategies were

PAGE 187

187 successful, others incorporated these recommendations into their ongoing health management practices Also important, post operati ve patients advise pre operative patients, which lends support to theories of influence and opinion leadership. Participants in this study also internalize experiences of similar others to feel good (or bad) about the ways in which they respond to their c ondition. Some feel that interaction with other members con firms the wisdom or fallacy of his or her own t reatment and recovery strategy. Experiential proximity then, provides the mechanism to mak e social comparisons as challenges of the condition are conf ronted. Members feel their peers understand the challe nges, which results in feelings of connection Further Theoretical Implications A noteworthy finding of this re search is that community members develop network ties with similar others despite the chall enges of reduced social cues. This fin ding resonates with social psychological research, wh ich has demonstrated that when people are under chronic physical stress, they are attracted to others with simila r experiences, and that often become s more important than structural similarity when forming relationships (Bandura, 1986). Thus the concept of experiential proximity described in thi s research is ground ed in the context of an online knee health community, yet has broader theoretical implications. This stud y has relevance for future studies of opinion leadership in online environments, and for those designed to investigate the dynamics of virtual communities, especially ones formed ar ound health topics. Peo ple are attracted to others of similar experience, a nd this appears to be more important than social similarity when they form new relationships online under various circumstances of physical (and likely mental) stress. This study demonstrates that members of an online community

PAGE 188

188 place considerable value on the ex change of information with peers; moreover, it extends previo us research by showing that participants value peer based information exchange in large part because of the experiential inf ormation available through relationships formed via textual inter action. Further this study supplements previous research regarding the content of experiential information; namely, that it is comprised of accrued practical wis dom and personal stories stemming from experience traumatic in many cases. T his research also documents a form of experiential information borne of the spec ificity of an online community focused on knee arthroplasty: it suggests, therefore, that content of experiential knowle dge may vary by setting. T he experiential k nowledge described in this stu dy is firmly grounded in the context of an online community formed around a potentially debilitating yet non life threatening condition, yet has broader theoretical implications. T his study argues that p eer based information exchange is generally an ongoi ng positive expe rience for those suffering a potentially incapacitating physical condition. People here feel rewarded in the sense that peers connect w ith them more deeply than others who are not faced with the physical and mental challenges this condition engenders. Online p eer based health information exchange fills an informa tional gap in the lives of those suffering distressing knee conditions. This research also contributes to literature examining disintermediation, or the capacity of the internet to allow the general public to bypass experts in its quest for information. It confirms that health related peer based information exchange continues to challenge the top down flow of information between doctors and patient s and

PAGE 189

189 demonstrates that the source o f influential information is not always a doctor, or even a health professional. This investigation adds to com munications literature by demonstrating how larger patterns of social structure (e.g., information gathering trust development, opinion leadersh ip, doctor /patient relation s ) are affected by a component of homophily and by technology. This combination helps shape unique forms of community that in turn shape so cial production While each online community represents a uni que set of technical features and a distinct user culture, this study adds insight into u nderstanding user behavior while adding theoretical significance to the relationships among experiential proximity, multi step flows of information, and online l eadership. Altho ugh previous resear ch has developed the con cept of experiential proximity, this study consider s how sharing experiential information may be linked to the appeal of network based pee r to peer information exchange and the development of opinion leaders in an online environment Recognition o f this homophilic component may facilitate more useful and persu asive online health information exchange. Limitations Although this research has several strengths, there are also limitations, both theoretical and practical. As there are a limited number of people who have expertise with this topic and would therefore be appropriate for this investigation, the study utilized non probability purposive sampling to research the phenomenon. In the absence of random s ampling, results then canno t be gene ralized to other online forums, or even to other knee health related communities. While a limitation, s ufficient description exists however, and future researchers will be able to determine if results here are transferable to other contexts. Also, findings of this study compare t o previous research,

PAGE 190

190 thus common characteristics can be n oted; future investigators may well find that results of this study resonate in other contexts. Practically, several unforeseen limitations surfaced as a consequence of the study design. Most notable was the narrow range in demographic characteristics of responding members. The sample was overwhelmingly Caucasian, female, educated, middle aged, and either retired or in occupations which encouraged physical inactivity. Further practical limitations are addressed below with respect to the data collection method employed. I n order to keep the data manageable in the observational phase of the study, message threads and the postings contained within those threads were ran domized and the median posting was extracted for analysis. This was necessary due to the very large number of postings that comprise the BoneSmart Knee Forum. This process did reveal certain members to be opinion leaders in the forum, however, others who may also be opinion leaders may not have surfaced due to random sampling. Although it is unlikely that an opinion leader more central to communications (via more and stronger ties to others) would have been revealed by sampling all threads and all posting s in this network, it is possible that a greater number of opinion leaders may have surfaced. Additionally, SNA as used here focuses on measures of centrality, yet some of the more complex analyses of network behaviors, such as sub groups and cliques (Was serman & Faust, 1994) or even advanced compo nents found in random graph modeling, could be utilized to provide more insight i nto the dynamics of networks such as this

PAGE 191

1 91 Other noteworthy limitations were due to a dependence on entirely voluntary responses. While the online questionnaire was available to all 2,481 Knee Forum members (approximately 700 are reportedly active at any one time), just 54 responded to the invitation and but 52 completed the questionnaire in its entirety, yielding a response rate of 7.5%. In an attempt to elicit responses on a number of wide ranging personal and theoretical orientations, the questionnaire may have been overly long and many opted not to reply. It is also possible that any number of members did not visit BoneSmart during the period the questionnaire was offered. Although the questionnaire yielded enough data to help answer the research questions posed in t his study, results would l ikely have been more robust had questionnaire response rate been greater. A related limitation is that not all respondents answered all questions offered in the questionnaire, thus missing data had to be taken into account This study also proposed to co nduct online interviews with 25 BoneSmart members to be selected from among those who provided an e mail address at the conclusion of the questionnaire. While 33 of the 52 questionnaire respondents did provide valid e mail addresses indicating they would be willing to answer open ended questions as part of the online interview process, only 17 responded to an introductory e mail initiatin g the interview process, and just 10 ultimately completed all interview questions. Those who did not complete all interv iew questions did, however, provide essential data for questions they completed. Despite the fact that the number of completed interviews was fewer than proposed, participating respondents did provide rich and detailed data that proved vital in answering t

PAGE 192

192 One solution to low response rates is to offer practical (rather than altruistic) incentives to increase rates. Recommendations Aside from offering solutions to the methodological problems explained above, recommendations fo r future research are in order. If the internet is assuming an increasingly vital role in global society, important for potential research would be to investigate in more detail, and with more qualitative tools of analysis, the ro le of this medium as a sur rogate substitute for important face to face social processes. Also, e xamination of online social network characteristics over time would illuminate the development of networks. Future research could explore similarities and differences among other online communities formed, like BoneSmart around people who also suffer serious yet non life threatening conditions, comparing for example, those who seek solutions to deteriorating eyesight with those who need hip replacement. Comparisons of similar communiti es would allow researchers to determine if opinion leadership, trust and credibility are formed in the same ways Are the characteristics and attributes of perceived opinion leaders in one community valued in another? Future research should further investi gate how people make use of online health forums, and f or which groups these forums hold p otential bene fit This should provide deeper insight into the effects of online forums within the total array of health related sup port. Ease of access combined with the positive effe cts these forums appear to have seems to justify such pursuit s. This research did not examine how individuals integrate online with offline information, such as that from friends and family, medical print sources or medical industry prom otional material (one questionnaire item did ask participants to rank order

PAGE 193

193 ten sources from a list) While this was not an objective, it does seem to be o f likely importance, thus future research should look more deeply into how these various sources are integrated, as well as the relationship of message source (and content) to outcomes. Scholarship would also benefit from further investigation into how computer and digital media intersect with the structure and functioning of social systems. A re virtual communities rebuilding social trust online, and have personal attributes become less relevant on the internet? This research would suggest so, yet Facebook and Twitter may have something to say about that. G iven the network nature of digital media communic ation, the social network approach is a useful way to address such questions, and this study has provided some initial directions for future research As the U. S. population ages, it becomes increasingly important to research how health information seeke rs use the web and draw on interactive and social web applications for health related purposes. What information, in its different forms, is retrieved altered, and output in this changing online environment and how is it further combined with information from other disparate sources in order to aid individual decision making? Web 2.0 the current i nternet evolution is associated with web applications that facilitate participatory information sharing interoperability user centered design and collaborat ion on the world wide w eb (D. J. Kim, Yue, Hall, & Gates, 2009). Web 2.0 has implications for health care in that it creates new levels of patient participation as well as unique and unprecedented opportunities for engaging people in their health, health care, and health research, and for connecting them with informal and

PAGE 194

194 formal caregivers, health professionals, and researchers. The emergence and adoption such as Google Health, with far reaching consequences for patient involvement, as centrality shifts away fr om health care providers as sole custodian s of medical data (Eysenbach, 2008). T he combination of two trends access to personal health records and social networki ng may lead to a powerful new generation of health applications where people share parts of their electronic health records with other consumers and With Web 2.0, d ifferent programmi ng languages, search functions, and applications that once stood alone (what some refer to as Web 1.0) are now increasi ngly (and effortlessly) integrated (Adams, 2010). For online communities such as BoneSmart, Web 2.0 technologies will allow members to cr eate and share quality video / audio and 3D simulations and integrate these efforts with older media such as film, television, knee health. Another crucial component of Web 2.0 is tha t it provides mechanisms to move this internet activity away from desktop computers to television s, laptops, tablets, cellular telephones and other mobile devices. These capabilities offer a richer user experience, and combined with the openness Web 2.0 pr ovides, should increase and enhance user participation. This is an area for future research, however. It should be noted that Web 2.0 also creates comple x privacy issues. Internet users perceiving information they p ost or disclose as ephem eral may be una ware that web information is often permanently archived and may be accessible long term (e

PAGE 195

195 g., by future employers). Little is known about actual user awareness of these privacy issues as they pertain to Web 2.0 technologies (Zimmer, 2008). This study co ntributes to what is known about health related online forums by examining in more detail what motivates people to make use of forums, and empirically link ing this to how effective people perceive these forums to be. T his study also extends previous resear ch by showing that peer based information exchange is valued because of the experiential information available through relationships formed online. Results indicate that how people use online health forums and what it is that attracts people to them is rel ated to how effective forums are believed to be in helping users cope with a particular health condition. These findings are important as they specify what it is that makes these forums conduits to impr oving quality of life for users

PAGE 196

196 A B Figure 5 1. A: Original two step flow of communication. B: Multi step flow of influential communication in an online knee health community. opinion leader m edia sources BoneSmart user decision i nterpersonal sources m edical professional experience knowledge 1 experiential experts Verification 2 mass media print radio

PAGE 197

197 APPENDIX A ONLINE QUE STIONNAIRE Shared Experience in a Virtual Community PART 1: GENERAL INTERNET USE 1. Do you access the internet mostly from: home work friend or family member's home public library other? please specify: For the following items, please enter a number from the scale to indicate how much your reasons for using the internet are like these reasons for using the internet : Not at all = 1 .... 2 .... 3 .... 4.... 5 = E xactly 2. I USE THE INTERNET : a) to encourage others b) to belong to a group c) when I have nothing better to do d) to gather points of view e) to express myself f) to do resea rch g) to tell others what they should do h) to escape from my daily routine i) because I enjoy answering questions j) to pay bills k) to get someone to do something for me l) to meet new people m) when I'm lonely n) to get news o) to download/watch movies p) to look for health information q) to shop for myself r) to download/listen to music s) to buy gifts t) to surf the web for fun u) to shop for others v) to play games w) other Please enter a number from the scale to indicate how much you agree or disagree with the following 3 statements: Strongly Disagree = 1 .... 2 .... 3 .... 4 .... 5 = Strongly Agree 3. I USE THE INTERNET :

PAGE 198

198 a) because it's the cheapest w ay to communicate with others b) to communicate with friends and family c) because it's the easiest way to communicate with others d) other 4. Number of hours online in a typical day NOT WORK RELATED PART 2: SEEKING KNEE RELATED HEALTH INFORMATION 5. How do you gather knee related health information? Please rank these items in order of use. Place a 1 in the box after the way you gather the most information, a 2 after the second most, and so forth. Place an X if you don't use this at all. Do not use any number more than once. a) newspaper b) medical professional c) magazines d) television e) radio f) internet BoneSmart g) internet other h) support group meetings i) friends j) family 6. Average number of hours spent online in a typic al WEEK looking for KNEE RELATED health information, NOT INCLUDING BONESMART COMMUNITY 7. Average number of hours spent online in a typ ical WEEK in BONESMART community, for any reason 8. How long have you been a member of the BoneSmart community? 3 months or less more than 3 months but less than 6 months more than 6 months but less than 1 year more than 1 year but less than 2 years 2 years or more 9. Choose ONE statement below that best describes your MAIN re ason for joining the BoneSmart community: newly diagnosed w/knee related health issue and looking for general info newly diagnosed and considering MIS robotically assisted partial knee replacement you've had MIS w/robotics and want to share info to gather info for others w/knee related health problems to find a surgeon best source for knee related health information to gather post op info (recovery or therapy)

PAGE 199

199 to meet people who may be in similar circumstances you're post op TKR and want to share info other? please specify: Please enter a number from the scale to indicate how much you agree or disagree with the following 2 statements: Strongly Disagree = 1 .... 2 .... 3 .... 4 .... 5 = Strongly Agree 10. This community is important 11. I consider myself involved in this community Sharing information is common in communities like BoneSmart. 12. If you participate in the BoneSmart community to gather information, aside from general postings, whom do you contact most often to get the knee related information you need? Enter one (1) screenname (same as "user name" or "handle"): 13. Not related to any specific topic, but in general, who are leaders in this community? Enter 3 screennames: 1) 2) 3) PART 3: THE ONLINE OTHER One element of computer communications is how people relate to each other in online environments. Thinking about others in general in this community, AND YOUR MAIN CONTACT in #12 above, please enter a number from the scale that reflects your level of agreement with each of the following 6 statements. Enter a NUMBER in EACH box: Strongly Disagree = 1 .... 2 .... 3 .... 4 .... 5 = Strongly Agree 14. Others /my main contact in this community are/is concerned about the ability of people to get along others my contact 15. Others/my main contact in this community are/is concerned about what is important to others others my contact 16. Others/my main contact in this community wi ll do everything they can to help others others my contact

PAGE 200

200 17. Others/my main contact in this community are/is responsive to my posts others my cont act 18. I can always count on getting a lot of responses to my posts from: others my contact 19. I can always count on getting responses to my posts fairly quickly from: others my contact Please enter a number from the following scale that reflects your level of agreement with each of the following 8 statements: Strongly Disagree = 1 .... 2 .... 3 .... 4 .... 5 = Strongly Agree 20. Posts in this community often c ontain personal information 21. Others in this community seem willing to divulge private information about themselves to others 22. I come to this community to get information about a particular topic 23. I come to this community to give others information I have about a particular subject 24. I come to this community to share my skills and abilities with others 25. I generally have faith in humanity 26. I feel people are generally reliable 27. I generally trust other people unless they give me a reason not to A factor in computer communications is what people THINK others are like even though they've never met in person and have only interacted through computers. Thinking again about OTHERS in general in this community, AND YOUR MAIN CONTACT in #12 above, please enter a number from the scale that reflects WHAT YOU THINK of others in this community. Please read items carefully and enter a NUMBER in EACH box: 28. Others/my main contact in this community are/is: Unreliable = 1 .... 2 .... 3 .... 4 .... 5 = Reliable others my contact 29. Others/my main contact in this community are/is:

PAGE 201

201 Uninformed = 1 .... 2 .... 3 .... 4 .... 5 = Informed others my contact 30. Others/my main contact in this community are/is: Inexpert = 1 .... 2 .... 3 .... 4 .... 5 = Expert others my contact 31. Others/my main contact in this community are/is: Dishonest = 1 .... 2 ... 3 .... 4 .... 5 = Honest others my contact 32. Others/my main contact in this community are/is: Unfriendly = 1 .... 2 .... 3 .... 4 .... 5 = Friendly others my contact 33. Others/my main contact in this community are/is: Unpleasant = 1 .... 2 .... 3 .... 4 .... 5 = Pleasant o thers my contact 34. Others/my main contact in this community are/is: Selfish = 1 .... 2 .... 3 .... 4 .... 5 = Unselfish others my contact 35. Others/my main contact in this community: Don't think like me = 1 .... 2 .... 3 .... 4 .... 5 = Think like me others my contact 36. Others/my main contact in this community are/is: Unlike me = 1 .... 2 .... 3 .... 4 .... 5 = Like me others my contact 37. Others/my main contact in this community have personalities: Different from mine = 1 .... 2 .... 3 .... 4 .... 5 = Similar to mine others my contact 38. Others/my main contact in this community: Don't share my beliefs = 1 .... 2 .... 3 .... 4 .... 5 = Share my beliefs others my contact 39. Others/my main contact in this community: Don't treat people like I do = 1 .... 2 .... 3 .... 4 .... 5 = Treat people like I do ot hers my contact 40. Others/my main contact in this community are/is from: A social class different from mine = 1 .... 2 .... 3 .... 4 .... 5 = A social class similar

PAGE 202

202 t o mine others my contact 41. Others/my main contact in this community are/is: Culturally different from m e = 1 .... 2 .... 3 .... 4 .... 5 = Culturally similar to me others my contact 42. Others/my main contact in this community have: Economic status different from mine = 1 .... 2 .... 3 .... 4 .... 5 = Economic status similar to mine others my contact 43. Others/my main conta ct in this community have: Backgrounds different from mine = 1 .... 2 .... 3 .... 4 .... 5 = Backgrounds similar to mine others my contact PART 4: ABOUT YOU 44. Sex ma le female prefer not to answer 45. Age on your last birthday 18 29 30 41 42 50 51 57 58 64 65+ prefer not to answer The next TWO items are for U.S. residents ONLY. If you reside in Canada, skip to question 48. If you reside in the U.K. skip to question 49. If you do not reside in the U.S, Canada or the U.K., skip to question 50. 46. Ethnicity American Indian/Alaskan Native Asian Black Hispanic Black non Hispanic Caucasian Hispanic Caucasian non Hispanic Hawaiian/Pacific Islander prefer not to answer other: 47. Personal annual income $U.S. $0 $20,000 20,001 40,000 40,001 60,000 60,001 80,000

PAGE 203

203 80,001 100,000 100,001 120,0 00 120,001+ prefer not to answer 48. Personal annual income $Cana dian C$0 C$20,606 20,607 41,212 41,213 61,824 61,825 82,424 82,425 103,030 103,031 123,624 123,625+ prefer not to answer UNITED KINGDOM RESIDENTS ONLY 49. Personal annual income £ £0 £13,200 13,201 26,400 26,401 39,6 00 39,601 52,800 52,801 66,660 66,661 79,200 79,201+ prefer not to answer 50. If you do not live in the U.S., Canada or the U.K., what country do you live in? 51. Are you employed? full time (40+ hours per week) part time not employed retired prefer not to answer 52. Occupation Executive, Administrative, Managerial Professional, Technical (example: engineering, law, educators, health care, arts, IT) Administrative Support (clerical, banking, postal, secretarial) S ales/Marketing Skilled Labor (example: construction trades, mechanical, drivers) Labor (example: construction labor, helpers, sanitation, factory assembly line, fast food) other: 53. Is your city or town population: 150,000+ 50,000 150,000 less than 50,000

PAGE 204

204 54. Last level of education completed: primary (elementary) secondary (high school) university (college) advanced degree (masters) advanced degree (doctorate) prefer not to answer other: We would like to ask follow up questions by e mail in the next few months about your personal experiences in gathering and acting on knee related health information. If you agree, please provide a valid e mail address for follow up. e mail address: Thank you very much. Shared Experience in a Virtual Community

PAGE 205

205 APPENDIX B INTRODUCTORY E MAILS Introductory e mail mail. These questions ar e not survey type questions, they are more in depth, asking you to think (and write) about your experience of being a member of an online forum and how you think this particular forum works for you. I'd first like to ask if had LTKR, RTKR, partial (and wh ich one), or both? And, if you would, provide the handle you used in BoneSmart. That would help make this research more accurate. Looking forward to your reply. Thanks, and Happy Holidays. Mic Brookshire Doctoral Candidate University of Florida Round 1 e mail Okay understanding how online communication works. Please think these over and take your time. There are no right and wrong answers, I want to know how you personally feel about these things. Here are four questions (three are multi part). I think three rounds would be best these four then three similar ones after these (finishing the nuts and bolts), then a few questions regarding your personal experiences having a knee undertaking this study (I do not, and never have, had a knee condition). If you have in the body of an e mail. Thank you very much. but 8 or 10 pages in the next day or two would be great. Just kidding. Anything you offer will help this research. Happ y Holidays, and thanks! MB (To those who did not respond) willing to answer a few questions after you completed the questionnaire in BoneSmart. I o help with this BoneSmart research. My deadline is fast approaching. There are three rounds of questions, these four, then three, then three more about your personal knee related experiences. Here's the first round. In any event, I wish you the best in the new year. MB

PAGE 206

206 APPENDIX C ONLINE INTERVIEW QUE STIONS Round 1 1: How do you decide that the information you get in BoneSmart is trustworthy and believable? 2: Opinion leadership happens when people need information or advice and turn to others with whom they have some connection and feel they can trust. Are there opinion leaders in this forum? How are they opinion leaders? In the questionnaire, you chose ____ as a leader. Why? Has any one person in BoneSmart influenced your decision to have (or not have) knee surgery of any type? 3: You may act on certain information without verifying it if that information comes from a parent, loved one, or doctor in face to face interaction. That same information gathered in a text only setting may (or may not) be interpreted differently. Do you check the in formation that comes through a w ebsite by way of a computer against other sources (other people medical journals, newspaper, w ebsites)? In other words, is information that comes through a computer from someone is it less trustworthy because the information comes t computer (the medium)? (Lots of people mistrust information available on the internet and there is potential for misleading information from online others). 4: Before computers, trust among individuals was formed when people interacted face to face (it still happens this way). Computers seem to be changing that a bit though. How is trust developed and maintained in this online community? For you, how does

PAGE 207

207 trust in others develop in a text only environment? Why would you trust someone you may never meet in person? Does trust for you develop immediately here? Thanks again! Round 2 5: De scribe the experience of being a member of an online forum how does going online and into BoneSmart fit into your life? Do you go there just for knee related 6: How important is the internet for you in dealing with your knee related condition? (Not BoneSmart particularly, but the internet in general). How would you go about finding information about your knee condition without the internet ? And more generally, how do you view the internet ? A tool? A place to be? A way of life? Something else? 7: Ho w do you use the knee related health information you gather in BoneSmart? Do you share it? Online and/or offline? Does it change how you interact wi th your health care provider(s).

PAGE 208

208 Condition Questions f your knee journey, but how you felt about the journey the emotions you felt about the physical changes and what trying to see if there are common elements in exper iencing this phenomenon. How is your life different from what it was prior to your knee problem(s)? What can you not do now that you could before your knee condition became a problem? Or, if you u realized you had a knee changes affected your emotional state along the way from problem recognition until now. Can you describe the anxiety you felt (if any) leading up to surgery? What were your thoughts about whether or not you would be able to resume your pre problem lifestyle? About whether or not you would be the same person. (I understand many people identify themselves through their physical activities Oh, I c Can you describe how you felt the moment you learned you might need knee replacement?

PAGE 209

209 APPENDIX D ONLINE INTERVIEW RES PONSES RQ1 : How do people describe the experie nce of being a member of an online community? How does a community member define his or her role in the community? G7 : I knew I'd be relatively young to the other members. At th is point in my recovery (pretty much recovered) I go mostly to see if there are any new members close to my age, and offer MY perspective on some things. F5 : Being a member of BoneSmart has been a large part of my life for the last eight months. It has been an extremely helpful and comforting place to be. I think the ratio of my visi ts is 70% for information and 30% to follow well and was genuinely concerned about their progress/problems. I think that is because I am from the other side of the world and unfami liar with the systems and drugs in USA & UK. W8 : recovery. habit now than a real desire to find new information. T11 : I'm retired, and so time is not a constraint. I'm on other forums, and use them purely for getting help and information But e moticons are ESSENTIAL, as we can make sure British humour can be seen by those abroad. B12 : I have a shortcut to BoneSmart on my desktop, th us making access to site easy. It is part of my daily routine to check the Staff Discussion and the New Posts. I am truly interested not only in knees, but in the general well being of members. On the topic of how a participant felt he or she fi t into the community, some described their role in terms of what they got out of being a community member, others lurkers, others explained their roles through expressions of what they felt they owed the community, and some felt th B12 replacement, that interacting on BoneSmart would be part of my ministry. I feel that I fit into the group, however each of us bring a different quality to the group (Mind you, I am always the round peg in the square whole).

PAGE 210

210 P4 : people. They are the ones who need the support. We grow to care about the others and hope for them to do well. W8 not big on having to continue chit chat once a message thread has gone stale. I do not feel that I fit in with the group of people who form the bac kbone of the forum. G7 : I don't think I do fit in all that well. It isn't just an age thing, because there are other 40 somethings there, but my tastes and interests are far diffent [ sic ] than many of the members there. A peripheral of RQ1 asked members w hat they perceived BoneSmart to be: Is it a community B12 each participant choses [sic] whether they care to be part of the group or part of the community. T11 : It is a community, and there a number of things which help it to be like this. The action of the moderators is key, where they are happy for us to temporarily zoom off topic, if we return to it soon. Even if we do we are only nudged back on course with a smile, if we don't actually meet and talk, there would always be a chance that someone will get hold of the wrong of the stick and take offence. F5 : I do think of BoneSmart as a place, a fascinating pl ace. I also believe it is a community, a changing community of people who are drawn to it and drawn into it for the period of their recovery and rehabilitation from major surgery. I think online friendships are formed and maintained for a while and then f ade away as recovery progresses. M9 to share (whether it be about my knee problems or even about a personal issue). I visit to check up on other forum members, leave comments or word are a community group. The closeness of the members and moderators make it feel like a family. It goes way beyond just discussions about knee or hip conditions it feels more like body/min d therapy. W8 : For me, BoneSmart is an online forum; it is not a community in the purest sense of that word. D13 : I believe there is a community at Bone Smart for some people but I think the majority are like myself and are transient visitors at a time of need.

PAGE 211

211 U14 : Yes it is a Place to Go for inform ation, comfort and camaraderie. Yes this is a community because we all care about each other. RQ2 : What significance does the internet have for these community members as it relates to his or her physical (knee related) condition? B12 : It was important to me because I feel that knowledge is power I wanted to know everything about the treatment, su rgery, prognosis, etc. It is W8 : I have done mor e than my fair share of web surfing looking for information first about osteoarthritis and then about knee replacement. The internet is a tool, a source of entertainment, a supreme waste of time G7 : I don't know how I did much of anything without the inte rnet honestly. At first, it was a distraction, and now the internet (and computers in general) are a way of life. D uring the physical therapy recovery process, reading other people's experiences and such made things seem easier P4 : The internet is my win dow to the world. If I had no computer, I would miss out on so much. Information. Friendship. Shopping. Before the internet I would buy a book with information, or go to the library. I would have had to just accept what my doctor said about my knee and right. Talking to friends who have had knee replacements helps, but when some of the things others go through My computer and the internet is such a great way to keep in touc or alone. I can find almost any information quickly about any question that pops into my head. Yes, the internet is a way of life. I get nuts when my internet and pacing and a lost feeling F5 : Using BoneSmart encouraged me to look elsewhere on the net for more information. I soon found that there was all I needed on BoneSmart. Without the internet I would have been entirely reliant on my surgeon and possibly t he library. It is a tool and a place to be, but not a way of life now. In the past it was much more of a way of life and indeed, my husband and I met on an internet dating site. M9 : The Internet plays a large role for me with my knee conditions. I do quite a bit of medical research from many different sites online. Without Internet I would have to go to the library very frequently and rely more Internet as a way of life for me, especia lly now, going through so much and not being able to get around easily.

PAGE 212

212 RQ3 : Does this online community generate opinion leaders? If so, how? S1 : I place great store in the qualifications of PRBX76 an orthopedic nurse, has had joint replacement, she has cr edibility. She weighs in on EVERYTHING and either validates it or tosses it out. PGJ192, PFH16, RGG189 give the impression of carefully considering each post, and weighing in with care and helpful comments and p ostings Th ey always contributed information that I had not previously known. They had been through things I was currently going through, had researched the solutions and offered them to others. G7 : PRBX76 claims to (and it seems to me that she's not exaggerating) have decades experience as an ortho nurse. In her signature, she states this with a disclaimer. She's not just spouting facts she's read in a book (and if she is, she's doing a real good job faking it). D13 : than opinion a nd would point you in the right direction to get the information that you may require to make an informed decision. PRBX76 and PRBN66 are nearly always available V3 : PRBX76 input has influenced my decision to seek a second opinion o n my particu lar knee problems. They regularly recommend that people become their own advocates and n ot accept everything that an OS [ orthopedic surgeon] something different. PRBX76 is definitely an opinion leader. To a lesser de gree, PRBN66 and PFH164. extensive nursing experience in the orthopedic area plus her own experience with a knee replacement allow her to have empathy with people experiencing problems after surgery. M9 : PGJ192 is very opinionated, I think, and se ems like the type of person who would tell anyone the truth no matter how much it might hurt or seem P4 : RAN40 come s across as someone who is sh arp and has a nice per sonality. He seems to be a good fam ily man and an honest person. open to us and has posted pictures of his family and himself in his position as a police captain I also believe what PRBX76 says, and she usually backs up her info with medical drawings E17 : PRS19 is great becau se she is a microbiologist, she understands my problems with MRSA [ Methicillin resistant Staphylococcus aureus a bacterium responsible for several difficult to treat infections in humans ].

PAGE 213

213 M9 : If he [ my orthopedic surgeon] does recommend surgery, then I will definitely discuss this on BoneSmart and listen to all the opinions given, especially T11 : PRBX76 is the real opinion leader, because of her history and experience. RQ4: How is trust developed and maintained in thi s online community? M9 : I trust people about the same whether its face to face, phone, or online. I trust people until I am given a reason to mistrust them. I think trust is developed immediately in this online community until there is reason to believe ot advice on BoneSmart, especially since most everyone is either going to have a hip or knee replacement or just had one. BoneSmart for my information. I do trust the people o n the forum for the most part since they have all been through similar surgeries and some of them also work in the medical field. B12 : I believe another thing about electronic communication is that when you build trust in the source. Such is BoneSmart Perhaps part of the mistrust comes from the News media constantly talking about scams etc D13 : I have noticed on Bone Smart that if information that is misleading or harmful is published it i s generally removed or disputed very quickly by one of the moderators. P4 : I think it takes quite a lot of reading what a person has written before you can decide if you can trust them or not. They mu st seem constant and reliable. They need to show empathy plu s intelligence in their posts. They need to let us know facts about themselves and their situation. A picture helps a lo t and background info is good. They need to s eem steady, honest, supportive and caring. ough some speak when they have no knowledge, just because they like to connect and see their words in print If you read enough, you can eventually tell if someone is a blow bag or a reliable source of information. H15 : I do value a person's opinion based on consistency and how they might be viewed by others on an internet discussion board. How they handle different questions asked of them and their patience in dealing with the repetition of, shall we s ay nonsensical type questions. That would and has earne d my respect. S1 : Information suggested on BoneSmart is trustworthy if it comes from an individual with a background in medicine or considerable experience with knee replacement themselves.

PAGE 214

214 G7 : Trust, for me, in an online setting is consistency. I didn't trust people I met online, at first. But as time went on, I found that there are people online who ARE what they say they ARE. I certainly am. I trust some of the things on BoneSmart (the informative stuff, not the social room stuff) because the experience itself is such an intense one that people aren't much inclined to lie about it. It s a shared experience. Like crash victims just knowing someone else is or has survived the same thing, it s a bonding experience. F5 : I have interacted with people online for more than ten years and believe D13 : I would say that this [ trust] will never totally replace face to face interaction and is never as assured as with personal contact. If you asked me would I loan someone money that I had an online relationship with and trust them to repay me, then the answer would be no. If you asked me if I would accept advice from the same person, then the answer would be yes. B12 : There are several similar messa ge boards but this one has the support of a health professional who has many years of experience in orthopedic surgery, but who has experienced total knee replacement. Primarily, commonality of Interests, ideas and experiences are important. Honestly carin g for the wellbeing of the person within the community Boundaries or Rules that are clear help to maintain the environment (Examples No bad language, no advertising, and no derailing threads (changing the topic). I was pleased to see that regardless of ho w stupid a answer. I weigh info presented: Does it match the expert information, is it reasonable, is it presented in an appropriate manner (trust me there are some real sleazes even on BoneSmart!!), is the person sincere, doe s the person have similar values, is the person receptive to help? I honestly like the fact that people post pictures of themselves or families, or even their pets or interests (does give the text world a human component) RQ5 : How is credibility accorded to sources and messages in this community? T11 : You [sic] so I can had never had a new knee, nor from someone who appears illiterate. P4 to face non medical professional people say any more than people who I know online. Some people online are repeating things they heard, not things they know a s fact. I t's just that I always feel I hav e to confirm and expand on it. S ometimes people tell you things they heard or which circulate the internet with no evidence. I trust the non medical things, such as good ideas for managing with a hurt leg.

PAGE 215

215 V3 : P ersonally, I initially checked a number of sites and spoke to others who had undergone the surgery. I found the BoneSmart group to mirror others experience on many issues. I f I see something that seems contrary, I m a range of sources H15 : I'm not saying I don't trust [ BoneSmart ] people' s opinion but I will always try and find some one else or another source or more to back up what info I'm being given, pro and con. The more sources I can get agreeing on a subject t hen the better I feel about my final decision. When I make that decision, it is usually very hard to sway me from it and I'll defend my decision using all th e info I have gathered on it. But I'm not so bullheaded that I will not change my mind if new info on the subject becomes available. Then I'll go through the process again. S1 : I mistrust information from all sources. I would trust information from my doctor over randomly gathered information on the Internet but I trust information from a reputable In ternet source. People are people, wherever they expound on their opinions or experience. Technically speaking BoneSmart is not reputable. I decided by myself that I trust what they say because I am an educated person and it's a decision I made. It is impor tant to keep in mind whether there is a benefit to the provider of the information. F5 : Before I found BoneSmart I already had good information from my surgeon and from other websites, however I did not actually know anyone who had had the surgery To hear that good information again from people in my situation was very comforting. the best! N10 : We assume that people on this type of communication forum are REALLY patients, or nurses, etc. wi th knowledge of real pat ients. Therefore their opinions or statement of their experiences helped us to be aware of what to expect, good and bad. RQ 6 : Is information gleaned from the community otherwise verified, or taken as fact? H15 : I had learned through all my quests for more info on any subject I was interested not to take any specific information I might ge t as the only correct answer. I have always sought as much info from as many different sources as I could reach in th e time I had available to me. I can't remember when I h ad ever made a decision based on one source no matter who or what organization I got it from. G7 : I have other health problems aside from the knee issue and in the search for a proper diagnosis from a less than sensitve [sic] doctor, I would research onlin e. What I did was take notes, compare them to my own

PAGE 216

216 experiences, read some other people's experiences and find the best way to approach the subject with the aforementioned "doctor ." I take it [ BoneSmart information ] with a bit of skepticism and then check things out for myself or make a note to ask my own doctor. That's how I see forums like BoneSmart we share a similar experience, but stuff that happened to the others is based on their own viewpoint. M9 : I do not think getting information from the Inter net is less trustworthy as long as other sources are researched as well. U14 : Repeat: verification, there are plenty of scammers and fools out there. N10 : We checked elsewhere for information too, many sources about knee surgeries. Internet media and peopl e who had had knee replacement V3 : The wealth of information that is available via the web is truly amazing. Search engines such as Google provide instant access to information it would have taken months to locate when one had to depend only on books or publications at a reference library. Verification also included interviewing other replacement patients and consulting doctors and health care professionals. D13 : I do not think that the information received by written form, whether on a forum, website, jo urnal etc is any more or less reliable than spoken information. I accept that not everyone will verify information received. RQ 7: How do community members use the health related information they gather in these online communities? N10 : It gave us some bas e to ask questions o f the doctor, it also helped us to choose a doctor based on his experience doing replacement surgery. We also were aware of various manufacturers and types of replacement materials, procedures, etc. so we were familiar with those things when the doctor started telling us about the procedure he would use, materials he would use, the hospital experience, the recovery treatment, and expected outcome. V3 : As a result of using BoneSmart I have become a much more vocal advocate of my own c oncerns when dealing with doctors and other health care professionals. Numerous times I have referred others to BoneSmart Some have visited the site. I believe I have influenced some to become more vocal about their concerns when discussing things with th eir doctors or health care professionals and we all owe that to the contributors to BoneSmart. P4 : I share the info with local people I know who are facing a knee replacement I gave one a great book which I found out about on

PAGE 217

217 BoneSmart and told another a bout that book and she immediately ordered it. I think the surgeon is glad when his patient knows anatomy and what the knee replacement consists of. I know they appreciate it when the patient can find answers to most of their simple questions elsewhere. Le ss unnecessary questions. B12 : I share the information mostly off line. I believe that the information I have gained helps me to ask appropriate questions and advocate for myself and my family U14 : YES, I have learned things that I have challenged my doc tor about, knowledge is very empowering G7 : I do share it. In fact, there are at least six people I've sent the [BoneSmart] link to who were going through a knee or hip replacement. [With my doctor] I know how to word things properly or ask things the ri ght way. M9 : I just shared some knee related information I discovered on BoneSmart with a neighbor, online with a few friends through Yahoo messenger, and my family at home. I also used some of the comments that PRBX76 had typed to me, when speaking to m y orthopedic doctor. It does change how I tried and know does not work for me. For instance, the doctor t ried to convince me yesterday to have a cortisone shot in my knee. I have had about 15 of these shots in the past and not one time have they helped me shot, when I know it will not hel members about speaking up for myself, I know I would have allowed the doctor to give me another useless shot. W8 : Before I had surgery, I was able to mine the BoneSmart site and put together a list of questions th at I asked my doctor. I have mentioned it to people who were facing knee replacement surgery. Revisiting the SNA observation component of this study (Figures 4 1, 4 2), and both questionnaire results (Tables 4 1, 4 2, 4 4) and participant interviews (prim arily responses to Research Question 3), it is obvious that PRBX76 fits the definition of role in the community, an e mail interview was conducted and her comments in res ponse to a separate set of questions were distilled.

PAGE 218

218 Why did you join this forum? PRBX76 : Simply because I had just learned I needed a replacement and though I knew knee replacement in detail from being a theatre [operating h about the recovery. So I started looking around and BoneSmart was the first forum I came upon. I was immediately struck and very concerned at the poor information members were passing around. y do you think others look to you for opinions and advice? PRBX76 : Well, basically, I have spent almost my entire life working in this skills and information. One of the joys in my work was to pass on this O ne of the great problems with researching anything on the net anyone on a forum can spout advice which is totally off the wall and even say they are qualified in that area. You may have noticed that a few members have the legal disclaimer in their signat ures. This is because BoneSmart has contacted them (and me) and asked them to verify their qualification by giving details of their training etc. In a forum like BoneSmart it is particularly dangerous to have 'know it alls' spouting off advice as if they w ere qualified in something and risk having nave or desperate members following it at risk of their well being and progress. Once a member joined who was in great pain with his knee. Doctor after doctor told him his knee was fine and he should j ust have more PT. This had been going on for about a year. He posted his latest xrays for me to see and I was able to tell him that one of his prostheses was loose. Why it was qu ite plain to me. So he saw one more surgeon who spotted it too, he had his revision and was perfectly fine after that! More recently a lady posted that her mother had been told her knee replacement was so badly infected the leg could only be either f used with a long metal rod or amputated! After some research I (and another) advised her to go to a certain surgeon in Texas who did some tests and announced she had no infection present, just a loose implant and did a revision for her. Whether this will b e successful only time will tell, but the chances seem pretty good. I have numerous other such examples, but you get the picture.

PAGE 219

219 PRBX76 : Well, I still am a nurse. I am still learning still practicing, in real know I take every opportunity to talk about joint replacement(and a few people coming to me and going away with information about hip/knee replacement and maybe even a recommendation to a surgeon, advice how are too young. I recently met one again, a manager who sprinted up some Seems I had advised him to get his hip done, told him how to manage his who was more delighted, him or me! timer, staff still regularly sought my input and advice on a variety of issues. But then, in my time, I had been a theatre [operating room] manager for many years and also had my own business running training courses for theatre staff on orthopaedic surgery. I have also written a text book which, though I could never get it formally published, printed them off at home and sold an awful lot of them across the UK and one or two other places. Probably around 300 or so. and what that means to you? PRBX76 : I get satisfaction from doing this, and it validates my sense of self worth. I express this through stories. I was in the theatre [operating room] one day where a hand operation was in progress. I noticed the houseman (junior doctor assisting) was wearing a gold chain around his neck. I mentioned it to him, adding that he m ust remove it as soon as the case was over. He bristled at this (he was an Arab chap and they famously get very indignant when instructed by a female!) and I explained what is was well known, neck adornments chafed skin scales off the neck which then get c aught in the plume of warm air that comes out of the gown and thereby such nonsense whereupon the senior surgeon admonished him saying e has more years of But being able to pass on information and skills to my colleagues was one thing; being able to care for and ensure the safety of the patients was the other and was never just a job to me. There was nothing I liked more than having the opportunity to go the extra mile for my pa tients In short, I love doing this. It gives me an outlet for my natural teaching skills, it enables to me help people in their journey, to face their fears and

PAGE 220

220 make good choices, to enable a less distressing recovery. There seems to be an attitude in the US that recovery from joint replacement, particularly knees, has to be a painful battle and the patients get abused and assaulted by PTs on a regular basis as a result. Emotional blackmail is rampant with thin 6 weeks your knee will set Aside from your personal experience and knowledge, where do you get the information you pass along in BoneSmart? How do you keep current with information that relates to your participation in BoneSmart? PRBX76 : I have a subscription to the Journal of Bone & Joint Surgery and attend conferences because I am a nurse. I also buy text books on a fairly regul ar basis. I have tried picking the brains of one or two doctors I know and whose opinion I trust, but they are understandably a little cagey about it. Trouble is, every surgeon has his own ideas about things and I would probably end up with as skewed an op inion as I started with. However, I think BoneSmart understands that there is a vast amount of knowledge and experience amongst our membership. We've had a chemist, a microbiologist and a technical engineer and we currently have a physician. But their real value is the experience they've had in their own replacements. That's a valuable resource. I also learn from them. In the works is a BoneSmarties Club where members of long standing who have consistently given good support and advice to other members will have a role in continuing to do that. Membership will be by application and closely vetted by the team. It should be a really interesting development of BS. Is there anything beyond a "knee condition" that makes this a group? Or is it just random p eople with the same condition on a message board? Most im portantly, is this a "community? PRBX76 : Initially people come to the forum because they have knee problems but it's a very open door community and newcomers are welcomed without hesitation. Other people see this and join because they sense the camaraderie in the group. We've had new members comment on this time and again in their very first post. Only yesterday one member said he joined 'because of the positive vibes coming out of this place'. Not only that but people bond. They notice if a member hasn't posted for a few days and worry about them when they are having problems or going for tests. A lot of our past members have a very tight community on Facebook, for instance. Others have made fr iends in real life and have travelled miles to

PAGE 221

221 meet each other. There is shortly going to be such a meeting in New York and the group are then travelling on to meet another member. I think you'd call that a community, don't you?

PAGE 222

222 LIST OF REFERENCES Abrams, L. C., Cross, R., Lesser, E., & Levin, D. Z. ( 2003 ) Nurturing interpersonal trust in knowledge sharing networks. Academy of Management Executive, 17 (4), 64 77. Adams, S. A. (2010). Revisiting the online health information reliability debate in the disciplinary literature and website review International Journal of Medical Informatics, 79 391 400 A lreck, P., & Settle, R. (2004). Survey research handbook (3 rd ed.). New York: McGraw Hill. Arndt, J. (1967 ). Role of product related c onversations in the diffusion of a new p rodu ct. Journal of Marketing Research 4 291 295. Atkinson, R. D. (2008, October). Why is the Digital Information Revolution So Powerful? Report from Information Technology and Innovation Foundation R etrieved December 12, 2008, from http://www.itif.org/index.php?id=34 Auerbach, C. F., & Silverstein, L. B. (2003). Qualitative data. New York: New York University Press. Aune R K ., & Kik uchi T. (1993). Effects of language intensity similarity on perceptions of credibility, relational attributes, and persuasion. Journal of Language and Social Psychology 12 224 237. Babbie, E. (2004). The practice of social research (10 th ed.). Belmont, CA: Wadsworth. B ales, R. F. ( 1951 ) Interaction process analysis: A method for the study of small g roups Cambridge, MA : Addison Westley Press, Inc. Balk, D. E., Tyson Rawson, K., & Colleti Wetel, J. (1993). Social support as an intervention with bere aved college students. Death Studies, 17, 427 450 Baller, R. D., & Richardson, K. K. (2009). The diffusion of suicidal thoughts. Journal of Health and Social Behavior, 50(3), 261 276. Bandura, A. (1986). Soc ial foundations of thought and action Englewood Cliffs, NJ: Prentice Hall. Baran, S. J., & Davis, D. K (2000). Mass communication theory Stamford, CT: Wadsworth. Basso, K. ( 1996 ). Wisdom sits in places: Notes on a Western Apache l andscape. In S. Feld & K. Basso (Eds.), Senses of Place (pp. 53 90). Santa Fe NM: School of American Research

PAGE 223

223 Benne, K. D. & Sheats P. ( 1948 ) Functional roles of group memb ers. Journal of Social Iss ues, 4 ( 2 ), 41 4 9. Benton C. ( 2008 September). Why broadband m atters B enton Foundation w ebsite Retrieved November 20, 2008, fr om http://benton.org/node/16820 Berger, C. R., & Chaffee, S. H. (Eds.). (1987). Handbook of communication science Newbury Park, CA: Sage Publications. B erlo D K Lemert J B Mertz R J. (1969). Dimensions f or evaluating the acceptability of message sources. Public Opinion Quarterly, 33 (4), 563 576. Bigley, G. A., & Pearce J. L. ( 1998 ) Straining for shared meaning in organization science: Problems of trust and distrust. Academy of Management Review, 23 (3) 405 4 42 Blanchard, A. L. (2008). Testing a model of sense of virtual community Computers in Human Behavior, 24 (5), 2107 2123. Blan ton, H. (2001). Evaluating the self in the c ontext of another : The three s elves model of social c omparison assimilation and c ontrast. In G. B. Moskowitz (Ed.), Cognitive Social Psychology: The Princeton Symposium on the Legacy and Future of Social Cognition (pp. 75 87). Mahwah, NJ: Lawrence Erlbaum. Blau, P. M. ( 19 64 ). Exchange and power in social life New York: J. Wiley & Sons. Boase, J. Horrigan, J. B., Wellman, B., & Rainie, L. (2006). The strength of internet ties. Pew Internet and Public Life Project. Retrieved April 20, 2009, from http://www.pew internet .org/PPF/r/172/report_display.asp BoneSmart.org staff. (2010). BoneSmart Hip / Knee Replacement Forum. Accessed May 21, 2010, from BoneSmart.org http://www.bonesmart.org/public_forum/index.php Book, C. L. (2008). Realizing the global promise of the internet. R eport from World Summit on Information Society Retrieved November 20, 2008, from http://www.elon.edu/docs/e web/predictions/IGF_RIO_SURVEY.pdf Borgatti S. P., Everett M., & Freeman, L. (1992, 1996, 2006). UCINET The program is distributed by Analytic Technologies Lexington, KY: http://www.analytictech.com/ucinet/ Borgatti, S. P., Mehra, A., Brass, D. J., & Labianca, G. (2009). Network a nalysis in the social s ciences Science, 323 892 895 Bourdieu, P., & Colemen J. S. (1991). Social theory for a changing society Boulder, CO: Westview.

PAGE 224

224 Bowker, N., & Tuffin, K. (2004). Using the online medium for discursive research about people with disabilities Social Science Computer Review, 22 (2), 228 241. Boyd, D. M., & Ellison, N. B. (2008). Social network sites: Definition, history, and s cholarship Journal of Computer Mediated Communication 13 (1), 210 230. Boyd, J (2003) The rhetorical construction of trust online. Communication Theory, 13 (4) 392 410 Brewer, M. B ., & Webber, J. G. (1991). Self e valuation effects of interpersonal versus intergroup s ocial c omparison. Journal of Personality and Social Psychology, 66 268 275. Broadway, S. C. (2005). Heal th information seeking behaviors on the internet among diabetic and healthy women Unpublished doctoral dissertation, University of Florida, Gainesville, FL. B rown, J. J., & Reingen P. H. (1987). Social ties and word of mouth referral behavior. Journal o f Consumer Research 14 350 3 62. Brown, J. O. Broderick, A. J., & Lee, N. (2007). Word of mouth communication within online communities: Conceptualizing the online social network. Jour nal of Interactive Marketing, 21 (3), 2 20. Bryman, A. (2001). Social research methods New York: Oxford University Press. Bull, M. (2001). The world according to sound: Investigating the w orld of Walkman New Media & Society 3 179 197. Burawoy M., Burton A., Ferguson A. A., Fox K. J., Gamson, J., Gartrell, N., et al. (Eds.). (1991). Ethnography unbound: Power and resistance in the modern metropolis Berkeley CA : University of California Press Burnett, G (2000 July ). Information ex change in virtual communities: A typology In formation Research, 5 (4). Retrieved Nov e mber 20, 2008, from http://InformationR.net/ir/5 4/ paper82.html Burns, E. (2008, December). Study: Social media s hot, e entertainment s n ot. ClickZ Retrieved July 4, 2009 from http://www.clickz.com/3632063 Burt, R. S. (1980). Models of network structure. In A. Inkeles, J. Coleman, & N. Smelser (Eds.), Annual Review of Sociology 6, 79 101. Burt, R. S. (1983). Range. In R. S Burt & M. J. Minor (Eds.), Applied Network Analysis (pp. 176 194). B everly H ills, CA: Sage Publications. Butler, J. K. ( 1991 ) Toward understanding and measuring conditions of trust: E volution of a conditions of trust i nventory. Journal of Management, 17 (3), 643 663.

PAGE 225

225 Caiani M., & Wagemann C. (2009). Online networks of the I tali an and G erman extreme right Information, Communication & Society, 12 ( 1 ), 66 109 Campbell, K., & Wright, K. B. (2002). On line support groups: An investigation of relationships among source credibility, dimensions of relational communication, and percepti ons of emotional support. Communication Research Reports 19 ( 2 ), 183 193 Carter D. ( 2004 ). Living in virtual communities: Making friends online. Journal of Urban Technology, 11 (3), 109 125. Cartwright D. & Harary, F. (1956). Structural balance: A generalization of Heider s theory Psychological Review, 63 277 293. Cartwright D. (Ed.). (1959). Studies in social power Ann Arbor, MI: Institute for Social Research. Cartwright, D., & Zander, A. (Eds.) (1968). Group dynamics: Research and theory New York: Harper & Row Publishers. Case, D., Johnson, J., Andrews, J., Allard, S., & Kelly, K. (2004). From two step flow to internet : The changing array of sources for genetics informa tion seeking. Journal of the American Society for Information Science and Technology, 55 ( 8 ), 660 669 Retrieved February 23, 2009, from http://www3.inters cience.wiley.com.lp.hscl.ufl.edu/cgi bin/fulltext/107617509/PDFSTART Chaffee, S. H. (1986). Mass media and interpersonal channels: Competitive, convergent, or complementary. In G. Gumpert & R. Cathcart (Eds.), Inter/media: Interpersonal communication in a media world (pp. 57 77). New York: Oxford University Press. Chan, K. K., & Misra, S. (1990). Characteristics of the o pinion leader: A new dimension. Journal of Advertising 19, 53 60. Clark, A. M. (1998). The qualitative quantitative debate: Moving from positivism and confrontation to post positivism and reconciliation. Journal of Advanced Nursing, 27, 1242 1249. Cline, R. J. & Haynes, K. M. (2001). Consumer health information seeking on the internet : The state of the art. Health Education Research, 16 671 692. Clouston, S. A. P., Verdery, A., Amin S., & Gauthier G. R. (2009). The structure of undergraduate association networks: A quantitative e thnography Connections 29 (2),18 31. Retrieved March 4, 2011, from http://www.insna.org/PDF/Connections/v29/2009_I 2_P 18 31.pdf

PAGE 226

226 Constant, D., Sproull, L., & Kiesler, S. (1996). The kindness of strangers: The usefulness of electronic weak ties for technical advice. Organization Scienc e, 7 (2), 119 135. Cooley, C. H. (1902). Human nature and the social order s ons Cooley, C. H. (1909). Social organization: A study of the larger mind New York: C. Scribner Corritorea, C. L., Krachera, B., & Wi edenbeck, S. (2003) On line trust: C oncepts, evolving themes, a m odel International Journal of Human Computer Studies, 58 737 758 Cre swell, J. W. (2008). Research design: Qualitative, quantitative, and mixed method approaches (3 rd ed.) Thousand Oaks, CA: Sage Publicatio ns. Cronkhite G & Liska J. (1976). A critique of factor analytic approaches to the study of credibility. Communication Monographs, 43 92 107. Crosby, L. A., Evans, K. R., & Cowles, D. ( 1990 ) Relationshi p quality in services selling: A n interpersonal influence perspective. Journal of Marketing, 54 68 81. Dahl R. A. (1957). The concept of power. Behavioural Science, 2 201 215. Davis, S. E. (2006). Identifying opinion leaders and elites: A longitudinal design. Library Trends, 55 (1), 140 157. Davison, K. P., Pennebaker, J. W., & Dickerson, S. S. (2000). Who talks: The social psychology of illness support groups. American Psychologist 55 (2), 205 217. DeFrances C J Lucas C A Buie V C & Golosinskiy A. (2008). National Hospital Discharge Su rvey. National health statistics reports, (5) Hyattsville, MD: National Delia J G. (1976). A c onstructivist analysis of the concept of credibility. The Quarterly Journal of Speech, 62 361 375. Demiris G. (2006). The diffusion of virtual communities i n health care: Concepts and challenges Patient Education and Counseling 62 178 188 Denzin, N. K. (1989). Interpretive biography Newbury Park, CA: Sage Publications, Inc. Denzin, N. K., & Lincoln, Y. S. (2005). The sage handbook of qualitative research (3 rd ed.). L ondon, UK: Sage Publications. Deutsch, M. (1958). Trust and suspicion. Journal of Conflict Resolution 2 265 279. Diani, M. (2003) Networks and social movements: A research programme In M. Diani & D. McAdam (Eds.), Social movements and ne tworks: Relational approaches to collective action (pp. 299 319). New York: Oxford Uni versity Press.

PAGE 227

227 Driskell, R., & Lyon ( 2002 ). Are virtual communities t rue communities? Examining the environments and elements of c ommunity. City & Community 1 373 39 0 Dwyer, C. ( 2007 qualitative s tudy. Paper presented at the 40 th Hawaii International Conference on System Sciences Retrieved December 18, 2007, from http://csis.pace.edu/~dwyer/research/DwyerHICSS 2007.pdf Eichhorn K. C. (2008). Soliciting and providing social support o ver the internet: An investigation of o nline eating disorder support gr oups Journal of Computer Mediated Communication, 14 67 78. Estabrook, L., Witt, E., & Rainie, L. (2007). Information searches that solve problems. Pew Internet & American Life Project Reports: Public Policy. Retrieved January 20, 2008, from http://www.pew internet .org/pdfs/Pew_UI_LibrariesReport.pdf Etzioni, A., & Etzioni O. ( 2001 ). Can virtual communities be r eal? In A. Etzioni (Ed.), The Monochrome Society (pp. 77 101). Princeton, NJ: Princeton Univers ity Press. Evans, D. S. (2008 May ). Special r eport: Broadband in America P c mag.com Retrieved December 2, 2008, from http://www.pcmag.com/article2/0,2817,2302118,00.asp Evans J. R., & Mathur, A. ( 2005). The value of online surveys. Internet Research, 15 (2), 195 219. Eveland, W. P. (2003) new communication technologies. Journal of Communication, 52 395 410 Eysenbach G. (2008). Medicine 2.0: Social networking, collaboration, participation, apomediation, and o penness Journal of Medical Internet Research, 10 (3), e22. Retrieved March 14, 2011, from http://w ww.ncbi.nlm.nih.gov/pmc/articles/PMC2626430 Fan, W., & Yan, Z. (2010). Factors affecting response rates o f the web survey: A systematic review. Computers in Human Behavior, 26 (2), 132 139. Fidel, R. (1993). Qualitative methods in information retrieval re search. Library & Information Science Research, 15 (3), 219 247. Filstead, W. J. (1979). Qualitative methods : A needed perspective in evaluation research. In T.D. Cook & C. S. Reichardt (Eds.), Qualitative and quantitative methods in evaluation research (p p. 33 48). Beverly Hills, CA: Sage Publications, Inc. Fischer, C. ( 1982 ). To dwell among friends Chicago, IL: University of Chicago Press

PAGE 228

228 Flaherty, L. M., Pearce, K. J., & Rubin, R. R. (1998). Internet and face to face communication: Not functional alt ernatives. Communication Quarterly, 46 (3), 250 268. Flick, U. (1998). An introduction to qualitative research Thousand Oaks, CA: Sage Publications. Foster, G. (1994). Fishing with the Net for research data. British Journal of Educational Technology, 25 ( 2), 91 97. Fox, K. (1991). The politics of prevention: Ethnographers combat AIDS among drug users. In M. Burawoy et al. (Eds), Ethnography unbound: Power and resistance in the modern metropolis (pp. 227 249). Berkeley, CA: University of California Press. Fox, S. (2005). Health information online. Pew Internet & American Life Project Retrieved July 15, 2009 from http://www.pew internet .org/PPF/r/156/report_display.asp Fox, S. (2006). Online health s earch 2006: Most internet users start at a search engine when looking for health information online Retrieved February 23, 2010, from http://www.pew i nternet .org/Reports/2006/Online Health Search 2006.aspx?r=1 Fox, S. (2008). The Engaged E Patient population. Pew Internet & American Life Project Retrieved July 15, 2009 from http://www.pew internet .org/~/media//Files/Reports/2008/ PIP_Health_Aug08.pdf.pdf Fox, S. (2010). All together now: The internet does not replace health professionals. Retrieved April 12, 2010, from http://www.pew internet .org/Expert s/~/link.aspx?_id=9E80C0EAB C1F4C07AA1E45D3AD4F4F82&_z=z Fox, S., & Jones, S. W. (2009). The social life of health i nformation. Pew Internet & American Life Project Retrieved July 2, 2009 from: http://www.pew internet .org/Reports/2009/8 The Social Life of Health Information.aspx?r=1 Freeman, L. C. (1979). Centrality on social networks. Social Networks, 1, 215 239. French, J. R., & Raven, B. (1968). Th e bases of social power. In D. Cartwright & A Zander (Eds.), Group dynamics: Research and theory (pp. 259 269). New York: Harper & Row. Friedkin, N. (1982). Information flow through strong and w eak ties in intraorganizational social networks. Social Netwo rks 3 273 285.

PAGE 229

229 Fung, R., & Lee, M. (1999). Ec trust (trust in electronic commerce): Exploring the antecedent f actors Proceedings of the Americas Conference of Information Systems 1999 Paper 179. Retrieved March 20, 2010, from http://aisel.aisnet.org/amcis1999/179 Galegher, J., Sproull, L., & Kiesler, S. (1998). Legitimacy, authority, and community in electronic support groups. Written Communication, 15, 493 530. Gals ton, W. ( 2000 ). Does internet str engthen c ommunity? National Civic Review, 89 (3), 193 202. Garton, L., Haythornthwaite, C. & Wellman, B (1997). Studying online social networks Journal of Computer Mediated Communication, 3 ( 1 ). Retrieved July 15, 2010, from: http://www3.interscience.wiley.com.lp.hscl.ufl.edu/cgibin/fulltext/120837719/HTML STARTW Gass R. H & Seiter J. S. (2002). Persuasion, social influence, and compliance gaining (2 n d ed. ). Boston, M A: Allyn & Bacon. Geertz, C. (1960). The religion of Java New York: Free Press. Geertz, C. (1965). The social history of an Indonesian town. Cambridge, MA: MIT Press. Gefen, D., & Ridings, C. ( 2002 ) Implementation team responsiveness an d user evaluation of cus tomer relationship management: A quasi experimental design study of social exchange theory. Journal of Management Information Systems 1 9 (1), 47 69. Giffin, K. (1967). The contribution of studies of source credibility to a theory of interpersonal trust in the communication process. Psychological Bulletin 68 (2), 104 120. Gillmor, D. (2006). We the media Gitlin, T. (1978). Media sociology: The dominant paradigm. Theory and Society 6 (2), 205 2 53. Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory Chicago: Aldine Press. Grannis, R. ( 1998 ). The importance of trivial streets: Residential streets and residential s egregation. American Journal of Sociology, 103 1530 1564. Granov etter, M. (1973). The strength of weak ties. American Journal of Sociology, 78 1360 1380.

PAGE 230

230 Granovetter, M. (1982). The strength of weak t ies: A network theory r evisited Sociological Theory, 1 201 233 Greene, J. C., & Caracelli, V. J. (2003). Making para digmatic sense of mixed methods practice. In A. Tashakkori & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioral research (pp. 9 1 110). Thousand Oaks, CA: Sage Publications. Grindrod, K. A., Marra, C. A., Colley, L., Cibere, J., Tsuyuki, R. T., Esdaile, J. M., et al. (2010). After patients are diagnosed with knee osteoarthritis, what do they do? Arthritis Care & Research, 62 (4), 510 515. Retrieved March 21, 2010, from http://www3.interscience. wiley.com.lp.hscl.ufl.edu/cgibin/fulltext/1233 37354/PDFS TAR Guba, E. G., & Lincoln, Y. S. (1994). Competing paradigms in qualitative research. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 105 117). Thousand Oaks, CA: Sage Publications, Inc. Gunther A. (1988). Attitud e extremity and trust in m edia. Journalism Quarterly, 65 279 287. Hamilton, M. B. (2009). Online survey response rates and times. SuperSurvey Ipathia, Inc. Retrieved August1, 2010 from http://www.supersurvey.com/papers/supersurvey_white_ paper_response_rates.pdf Hanneman, R. A., & Riddle, M. (2005). Introduction to social network methods Riverside, CA: http://f aculty.ucr.edu/~hanneman/ Hanson W. E., Creswell, J. W., Plano Clark, V. L., Petska, K. S., & Creswell J. D. (2005). Mixed methods research d es igns in counseling p sychology Journal of Counseling Psychology, 52, 224 235 Haythornthwaite, C. (2000). Onl ine personal networks: Size, composition and media use among distance learners. New Media & Society, 2, 195 226. Haythornthwaite, C. A. (2002) Strong, weak and latent ties and the impact of n ew m edia The Information Society 18 (5), 385 401 Heidegger M (1977). Basic writings. (D. Krell, Trans.). New York: Harper & Row. Hesse, B. W., Moser, R. P., & Rutten, L. J. (2010). Surveys of physicians and electronic health information. New England Journal of Medicine, 362 (9), 859 860. Hesse Biber S. N., & Lea vy, P. (Eds.). (2003). Approaches to qualitative research: A reader on theory and practice New York: Oxf ord University Press.

PAGE 231

231 Hiltz, S. R. ( 1984 ). Online communities: A case study of the office of the future Norwood, NJ: Ablex Pub lishing Corporation. Hin e, C. (2000) Virtual e thnography Thousand Oaks, CA: Sage Publications. Hodgson, S. (2004). Cutting through the silence: A sociological construction of self injury. Sociological Inquiry, 74 (2), 162 179. Hodkinson, P. (2002) Goth. I dentity, style and subcu lture Oxford, UK: Berg Publishers. Homans, G. C. (1950). The human group New York: Harcourt, Brace. Horrigan, J. ( 2008 ). Home broadband a doption 2008 Pew Internet and American Life Project Reports: Technology. Retrieved November 20, 2008, from http://www.pew internet .org/pdfs/PIP_Broadband_2008.pdf Hovland, C. (1951). Changes in attitude through communication. Journal of Abnormal and Social Psychology, 46 424 437. Hovland C. I. Janis I L & Kelley J J. (1953). Communication and persuasion: Psychological studies of opinion change New Haven, CT: Yale University Press. Hybye, M. T., Johansen, C., & Tjrnhj Thomsen T. (2005), Online interaction: Effects of storytelling in an internet breast cancer support group Psycho Oncology, 14, 211 220. Huffaker, D. A. (2009). Dimensions of leadership and social influence in online communities. Unpublished doctoral dissertation, Northwestern University, Evanston, IL. Hurst, L. (1991). M r. Henry makes a deal. In M. Burawoy et al. (Eds), Ethnography unbound: Power and resistance in the modern metropolis (pp. 183 202). Berkeley, CA: University of California Press. Internet World Stats (2010 ). Internet usage stats and population statistics Retrieved May 1, 2010, from http://www. internet worldstats.com/stats14.htm Jarvenpaa, S. L., Knoll, K., Leidner, D. E. ( 1998 ) Is anybody out there? Antecedents of trust in global virtual te ams. Journal of Management Information Systems 14 (4), 29 64. Jenkins, H. (2006). Convergence culture: Where old and new media collide New York: New York University Press.

PAGE 232

232 Jin, Y., Bloch, P. & Cameron, G. T. (2002). A comparative study: Does the word of mouth communications and opinion l ea dership model fit e pinions on the internet ? Proceedings of the Hawaii International Conference on Social Sciences. Retrieved February 12, 2010, from Co mmunications%20and%20Opinion%20Leadership%20Model%20Fit%20Epinio ns%20on%20the%20 internet .pdf Johnson, R. B., & Onwuegbuzie A. J. (2004). Mixed methods research: A research p aradigm whose time has c ome Educational Researcher, 33 ( 7), 14 26 Jones, Q. ( 1997 ) Virtual communities, virtual settlements & cyber archaeology: A theoretical outline Journal of Computer Mediated Communication 3 ( 3 ). Retrieved October 17, 2009 from http://jcmc.indiana.edu/vol3/issue3/jones.html Jones, R., & Ortlieb, M (2006). Online place and person making: Matters of the heart and self expression. Proceedings of the Ethnographic Praxis In Industry Conference 2006 (1) 214 228. Jones, S. G. (Ed.). (1995). In Understanding community in the information age (pp 10 35). Thousand Oaks, CA: Sage. Jsang A., Ismail, R., & Boyd C. (2007). A survey of trust and reputation systems for online service provision Decision Support Systems, 43 618 644 Karchmer, R. A. (2001). The journey ahead: Thirteen teachers report how the internet influences literacy and literacy instruction in their K 12 classrooms. Reading Research Quarterly, 36 (4), 442 466. Katz, E. (1957). The two step flow of communication: An up to date report on an hyp othesis. Public Opinion Quarterly, 21 (1) 61 78. Katz, E., & Lazersfeld, P. F. (1955). Personal influence: The part played by people in the flow of mass communications. Glencoe, IL: The Free Press. Kavanaugh, A. L., Reese, D. D., Carroll, J. M., & Rosson, M. B. (2005). Weak ties in networked communities. The Information Society, 21 (2), 119 131. Kaye, B., & Johnson, T. (2003). From here to obscurity?: Media substitution theory and traditional media in an on line world Journal of the American Society for Inf ormation Science and Technology, 54 (3), 260 273. Kennedy H. (2006). Beyond anonymity, or future directions for internet identity research New Media & Society 8, 859 875 Kim, D. J., Yue, K., Hall, S. P. & Gates, T. (2009) Global diffusion of the intern et XV. Web 2.0 technologies, principles, and applications: A conceptual framework from technology push and demand pull perspective. Communications

PAGE 233

233 of the Association for Information Systems 24. Retrieved March 16, 2011, from http://aisel.aisnet.org/cais/vol24/iss1/38 Kim, D. K. (2008). A guideline to design and conduct opinion leader based diffusion strategy for communicative health interventions: Methods of identifying opinion leaders and commu nication n etworks Paper presented at the NCA 94 th Annual Convention, San Diego, CA Nov 20, 2008 Retrieved January 24, 2010, from http://www.allacademic. C om/meta/p255677_index.html Kim, D. K., Chitnis, K., Vasanti P. N., & Singhal A. (2007). Opinion leadership in Indian villages and diffusion of E c houpal Journal of Creative Communications, 2, 345 360 King, C. W., & Summers, J. O. (1970). Overlap of opinion leadership across consumer product categories. Journal of Marketing Research, 7 43 50. Kious is S. (2001). Public trust or mistrust? Perceptions of media credibility in the information age. Mass Communication & Society, 4 (4), 381 403. Klapper, J. (1960). The effects of the mass media Glencoe, IL: Free Press of Glencoe. Kohler, W. (1927). The me ntality of apes New York: Harcourt, Brace & Company, Inc. Kollack, P., & Smith, M. A. (Eds.). (1999). Communities in cyberspace (pp. 3 25). London: Routledge. Kozinets, R. (1999). E tribalized marketing?: The strategic implications of virtual c ommunitie s of c onsumption European Management Journal 17 252 264 Kozinets, R (2002) The field behind the screen: Using netnography for marketing research in online c ommunities Journal of Marketing Resear ch 39, 61 72 Kurtz S Ong K Lau E Mowat F & Halpern M. (2007). Projections of primary and revision hip and knee arthroplasty in the United St ates from 2005 to 2030. Journal of Bone Joint Surgery (American), 89 (4), 780 78 5 Kvale, S. (1996). Interviews: An introduction to qualitative research int erviewing Thousand Oaks, CA: Sage Publications. Lasswell, H., & Kaplan, A. (1950). Power and society: A framework for political inquiry New Haven, CT: Yale University press. Lazarsfeld, P. F., Berelson, B., & Gaudet H. (1948). The people s choice: How the voter makes up his mind in a presidential election. New York: Columbia University Press.

PAGE 234

234 Lazarsfeld, P. F., & Merton, R. K. (1954). Friendship as a social process: a substantive and methodological analysis. In M. Berger, T. Abel, & C. Page (Eds.), Free dom and control in modern society (pp. 18 66). New York: Litton Publishing Lea, M., & Spears, R. (1992). Paralanguage and social perception in computer mediated communication. Journal of Organizational Computing, 2 321 341. Lefebvre, H. (1991). The produ ction of space. (D. Nicholson Smith, T rans. ). Oxford: Blackwell (Original work published 1974). Lehu, J M. (2004). Back to life! Why brands grow old and sometimes die and what managers then do: An exploratory qualitative research put into the French conte xt. Journal of Marketing Communications, 10 133 152. Levine, J. H. (1972). The sphere of influence. American Sociological Review 37 (1), 14 27. Lewicki, R J., & Bunker, B. B. ( 1995 ). Trust in relationships: A model of development and decline. In B. B. Bunker & J. Z. Rubin (Eds.), Conflict, cooperation, and justice: Essays inspired by the work of Morton Deutsch (pp. 133 173). San Francisco, CA: Jossey Bass. Lewin, K. (1936). Principles of topological psychology New York: McGraw Hill. Lewin, K. (1948). Resolving social conflicts. New York: Harper & Row Publishers. Lewin, K., & Grabbe, P. (1948). Conduct, knowledge, and acceptance of new values. Journal of Social Issues, 1 (3), 53 64. Lewis, D., & Weigert, A. ( 1985 ) Trust as a social reality. Social Forces 63 (4), 967 985. Lewis, T. (2006). Seeking health information on the internet: lifestyle choice or bad attack of cyberchondria? Media, Culture, & Soci ety, 28 (4), 521 539. Lieberman, M. A., Wizlenberg, A., Golant, M., & Di Minno, M. (2005). The impact of group composition on internet support groups: Homogeneous versus groups. Group Dynamics: Theory, Research, and Practice, 9, 23 9 250. Lin, N. (1973). The study of human c ommunication Indianapolis IN : Bobbs Merrill Company. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry Beverly Hills, CA: Sage Publications. Lindlof T. R., & Taylor B. C. (2002). Qualitative Communi cation Research Methods (2 nd ed.). Thousand Oaks, CA: Sage.

PAGE 235

235 Lorrain, F. P., & White, H. C. (1971). Structural equivalence of individuals in social networks. Journal of Mathematical Sociology, 1 49 80. Losina, E., Walensky, R. P., Kessler, C. L., Emrani, P. S., Reichmann, W. M., Wright, E. A., et. al. (2009). Cost effec tiveness of total knee a rt hroplasty in the United States: Patient risk and hospital v olume (Electronic version). Archives of Internal Medicine 169 (12), 1113 1121. Retrieved March 20, 2010, from http://archinte.amaassn.org.lp.hscl.ufl.edu/ cgi/content/full/169/12/1113 Lowery, S. A., & DeFleur, M. L. (1995). Milestones in mass communication research (3 rd ed.). White Plains, NY: Longman Publishers. Lders, M. (2008). Conceptualizing persona l media New Media & Societ y, 10 683 702 Luhmann, N. ( 1979 ). Trust and power N ew York: J. Wiley & Sons. Lyon, L. ( 1999 ). The community in urban society Prospect Heights, IL: Waveland Press, Inc. Maloney Krichmar, D., & P reece J. (2005). A multileve l analysis of sociability, u sability, and community dynamics in an o nline health c ommunity ACM Transactions on Computer Human Interaction, 12 (2), 1 32. Mann, C., & Stewart, F. (2000). Internet communication and qualitative research: A handbook for researc hing online. London UK : Sage Publications. Markham, A. ( 1998 ). Life online: Researching real experience in virtual space. Walnut Creek, CA: Altmira Press Marsden, P. ( 1988 ) Homogeneity in confiding relations. Social Networks 10 (1), 57 76. Marsden, P V., & Lin, N. (Eds.). (1982). Social structure and network analysis Beverly Hills, CA: Sage Publications. Martin K (2008, September 30). Ke ynote r emarks of Chairman Kevin Martin of the U.S. Federal Communications Commission Network Neutrality Confer ence: Implications for Innovation and Business Online, Copenhagen, Denmark Retrieved November 20, 2008, from http://www.fcc.gov/commissioners/martin/speeches2008.html Mayer, R. C., Davis, J. H., & Schoorman, F. D. ( 1995 ) An integrative model of organizational trust. Academy of Management Review, 2 0 (3), 709 734 McCarty, C. & Bernard H. R. ( 2003 ). Social network a nalysis In K. Christensen & D. Levinson (Eds.), Encyclopedia of co mmunity: From the village to the virtual world (pp. 1321 1325). Thousand Oaks, CA: Sage Publications.

PAGE 236

236 McCoyd, J. L. & Kerson, T. S. (2006). Conducting intensive interviews using email: A serendipitous comparative opportunity Qualitative Social Work 5 389 405. McCroskey J. C. (1966). Scales for the measurement of ethos. Speech Monographs, 33 65 72. McCroskey, J. C, & Richmond, V. P (1996). Fundamentals of human communication: An interpersonal perspective. Prospect Heights, IL: Waveland Press. McCroskey, J. C., Richmond, V. P., & Daly, J. A. (1975). The development of a measure of perceived homophily in interpersonal communication. Human Communication Research, 1 (4), 323 332. McCroskey J C & Young J T. (1981). Ethos and credib ility: T he construct and its measurement after three decades. Central States Speech J ournal, 32 24 34. McCroskey, L. L., McCroskey, J. C., & Richmond, V. P. (2006). Analysis and improvement of the measurement of interpersonal attraction and h omophily Com munication Quarterly, 54 (1), 1 31 McGrath, C., Blythe, J., & Krackhardt, D. (1996). Seeing groups in graph layouts. Connections, 1 9 (2), 22 29. Retrieved August 2, 2010, from: http://www.insna.org/PDF/C onnections/v19/1996_I 2.pdf McGrath, M. G. (2009). Centrality and power among website users: A social network analysis application. Paper presented at the 2009 International Conference on Computational Science and Engineering Retrieved January 31, 2020, from http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?asf_arn=null&asf_iid=null&asf_ pun5 282954&asf_in=null&asf_rpp=null&asf_iv=null&asf_sp=null&asf_pn=11 McKinlay, J., & Marceau L. (2008). When there is no doctor: Reasons for the disappearance of primary care physicians in the US during the early 21 st century Social Science & Medicine 67 ( 10) 1481 1491. McKnight, D. H., Choudhury V., & Kacmar C. (2002). Developing and validating trust measures for e commerce: An integrative t ypology Information Systems Research 13 (3), 334 359 McQuail, D. (2000). mass communication theory (4 t h ed.). Thousand Oaks, CA: Sage Publications. Mead. G. H. (1934). Mind, self and society Chicago: University of Chicago Press. Meho, L. I. (2006). E mail interviewing in qualitative research : A methodological discussion. Journal of the American Society fo r Information Science and Technology, 57 (10), 1284 1295.

PAGE 237

237 Meho, L. I., & Tibbo, H. R. (2003). Modeling the information seeking behavior of social Journal of the American Society for Information Science and Technology, 54 (6), 570 587. Mertens, D. M. (2003). Mixed methods and the politics of human research: The transformative emancipatory perspective. In A. Tashakkori & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioral research (pp. 135 164). Thousand Oa ks, CA: Sage. Merton, R. K. (1949). Patterns of influence. In P. Lazarsfeld & F. Stanton (Eds.), Communication Research (pp. 180 219). New York: Harper and Brothers. Meyerson, D., Weick, K. E., & Kramer, R. M. (1996). Swift trust and temporary groups. In R. M. Kramer & T. R. Tyler (E ds. ), Trust in o rganizations: Frontiers of theory and research (pp. 166 195). Thousand Oaks, CA: Sage Publications. Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook. Thousa nd Oaks, CA: Sage Publications. Mitchell, J. C. (Ed.). (1969). Social networks in urban situations: Analyses of personal relationships in Central African towns. Manchester, UK: Manchester United Press. Molotch, H., with Freudenburg, W., & Paulsen K. ( 2000 ). History r epeats itself, but how? City character, urban tradition, and the accomplishment of p lace. American Sociological Review, 65, 791 823. Moreno, J. ( 1934 ) Who shall survive? New York: Beacon Press. Morse, J. M. (1994). Designing funded qualitative r esearch. In N. K. Denzin & Y. S. Lincoln (E ds. ), Handboo k of q ualitative r esearch (p p. 220 235 ) Thousand Oaks, CA: Sage Publications. Muniz, A. M. Jr., & O Guinn T. C. (2001). Br and Community. Jou rnal of Consumer Research. Murray, C. D. (2004). An interpretive ph enomenological analysis of the embodiment of artificial limbs. Disability and Rehabilitation, 26 (16), 963 973. Murray, C. D., & Harrison, B. (2004). The meaning and experience of being a stroke survivor: An interpretive pheno menological analysis. Disabilit y and Rehabilitation, 26 (13), 808 816. Murray, C. D., & Sixsmith, J. (1998). E mail: A qualitative research medium for interviewing? International Journal of Social Research Methodology, 1 (2), 103 121 Nadel, S. F. (1957). The theory of social structure G lencoe, Ill: Free Press.

PAGE 238

238 Naquin, C. E., & Paulson G. D. (2003). Online bargaining and interpersonal t rust Journal of Applied Psychology, 88 ( 1 ) 113 120 Nisbet, M., & Kotcher, J. (2009). A two step flow of influence? Opinion leader campaigns on climate c hange. Science Communication 30 328 354. O Keefe, D. J. (1990). Persuasion: Theory and research. Newbury Park, CA: Sage Publications. Olivero, N., & Lunt, P. (2004). Privacy versus willingness to disclose in e commerce exchanges: The effect of risk awar eness on the relative role of trust and control. Journal of Economic Psychology, 25 (2), 243 262. Papacharissi, Z., & Rubin, A. M. (2000). Predictors of internet use. Journal of Broadcasting & Electronic Media, 44 (2), 175 196. Park, R. ( 1936 ). Human ecolog y. American Journal of Sociology, 17 (1), 1 15. Parks, M. R., & Floyd, K. (1996). Making friends in c yberspace. Journal of Communication, 46 80 97. Patton, M. Q. (2002). Qualitative research and evaluation methods (3 rd ed.). Thousand Oaks, CA: Sage Publica tions. Pettigrew, K. E., Fidel, R., & Bruce, H. (2001). Conceptual frameworks in information behavior. Annual Review of Information Science and Technology, 35 43 78. Pew Internet and American Life Project. ( 2 006). Online health s earch 2006: Most internet users start at a search engine when looking for health information online Retrieved May 1, 2010, from http://www.pew internet .org/Reports/2006/Online Health Search 2006. aspx?r=1 Pew Internet & American Life Project ( 2008 ). Demographics of internet user s Retrieved March 10, 2008 from http://www.pew internet .org/trends.asp Pew Internet and American Life Project. (2009). A shifting landscape : The social life of health information 2009 Retrieved Decem ber 12, 2009, from http://www.pew internet .org/Reports/ 2009/8 The Social Life of Health Information/02 A Shifting Landscape/3 Postma n, N. (1992). Technopoly New York: Vintage Preece J. (1999). Empathic communities: B alancing emotional and factual commu nication Interacting with Computers, 12 63 77 Preece J. (2000) Online communities, designing usability, supporting sociability Chicester,West Sussex, UK: J. Wiley & Sons.

PAGE 239

239 Preece, J., & Ghozati, K. (2001). Experiencing empathy online. In R. E. Rice & J. E. Katz (Eds.), The internet and health communication (pp. 237 260). Thousand Oaks, CA: Sage Publications. Prell, C. L. (2003). Community networking and social capital: Early investigations. Journal of Computer Mediated Communications, 8 (3) (Electronic version). Retrieved July 1, 2010, from http://www3.interscience.wiley.com.lp.hscl.ufl.edu/cgi bin/fulltext/120837878/ htmlstart Quinn, N. (1982). Commitment in American marriage: A cultural a nalysis American Ethnologist, 9 ( 4 ), 775 798. Rafaeli, S. & Raban D. R. (2005). Information sharing online: A research challenge. International Journal of Knowledge and Learning 1, 62 79. Rainie, L., & Horrigan, J. (2002). The broadband d ifference : behavior changes with high speed internet connections at home Pew Internet & American Life Project Reports: Broadband. Retrieved November 20, 2008, from http://www.pew internet .org/pdfs/PIP_Broadband_Report.pdf Rainie L. & Horrigan, J. (2005) Internet: Th e mainstreaming of online life. Pew Internet & American Life Project Reports: Internet evolution. Retrieved November 20, 2007 from http://www.pew internet .org/pdfs/ internet _Status_2005.pdf Rains, S. A. (2007) Perceptions of traditional information sources and u se of the world wide web to s eek health information: Findings f rom the Health Info rmation National Trends Survey. Journal of Health Commun ication, 12 (7) 667 680 Rapoport, A. ( 1952 ). Ignition phenomena in random nets Bulletin of Mathematical Biophysics,14, 35 44 Reicher S Spears R & Postmes T. ( 1995 ) A social identity model of deindividuation phenomena. European Review of Social P sychology 6 161 198 Rhee, J. W., Kim, E. M., & Kim, H (2007). Exploring o nline o pinion leadership: A validity test of the concept in the digital a ge. Paper presented at the International Communication Association Annual Meeting, San Francisco, CA Ma y 26, 2007. Retrieved January 12, 2010 from: http://www.allacademic.com//meta/p_mla_apa_research_citation /1/7/3/1/4/pages 173140/p1731401.php Rheingold, H. ( 1993 ). The virtual community: Homesteading on the electronic frontier. London: MIT Press. Rheingo ld, H. (1994). The virtual community: Finding connection in a ehavior zed world. London: Secker & Warburg

PAGE 240

240 Rice, R. E., & Katz, J. E. (2001). The internet and health communication: Experiences and expectations. Thousand Oaks, CA: Sage Publications. Richar dson, M. ( 2007 ). Being in the market versus being in the plaza: Material culture and the construction of social r eality in Spanish America. In S. Low & D. Lawrence Zuniga (Eds.), The a nthropology o f space and p lace (p p. 74 91 ) Malden, MA: Blackwell Ridin gs, C. M., & Gefen, D. (2004). Virtual community attraction: Why people hang out online. Journal of Computer Mediated Communication, 10(1). Retrieved November 25, 2010, from http://onlinelibrary.wiley.com/doi/10.1111/j.1083 6101.2004.tb00229.x/full Ridings, C. M., Gefen, D., & Arinze, B. (2002). Some antecedents and effects of trust in virtual communities. Journal of Strategic Information Systems, 1 1, 271 2 95 Right health.com. (2010). RightHealth results for knees. Retrieved April, 16, 2010, from http://www.righthealth.com/Health/Knees s?lid=goog ads sb 8536643334 Rintels, J. (2008, December ). An action plan for America: Using technology and innovation to address our nation s critical c hallenges: Telehealth. Benton Foundation w ebsite Retrieved May 1, 2010, from http://www.benton.org/initiatives/broadband_benefits/ action_plan/health _care#telehealth Rogers, E. M. (1962). Diffusion of innovations New York: Free Press of Glencoe. Rogers, E. M. (1995). Diffusion networks. In Diffusion of innovations, (pp. 281 334). New York: Free Press. Rogers, E. M. ( 2001 ). The d igit al d ivide. Conver gence, 7 (4) 96 111 Rogers, E. M., & Cartano, D. G. (1962). Methods of measuring opinion leadership. Public Opinion Quarterly, 26 435 441. Roethlisberger F. J. & Dickson, W. J. (1939). Management and the worker New York: Routledge. Rotter, J. B. ( 197 1 ). Generalized expectancies for interpersonal trust. American Psychologist, 26, 443 452. Rubin, H. J., & Rubin, I. S. (1995). Qualitative interviewing: The art of hearing data Newbury Park, CA: Sage Publications, Inc. R yan G. W., & Bernard, H. R. (2003 ). Techniques to identify t hemes Field Methods 15 (1), 85 109

PAGE 241

241 Sabel, C. F. ( 1993 ) Studied trust: building new forms of cooperation in a volatile economy. Human Relations 46 (9), 1133 1170 Sassenberg, K. (2002). Common bond and common identity groups on the internet: Attachment and normative behavior in on topic and off topic chats. Group D ynamics: Theory, Research and P ractice, 6 27 37. Schneider, S. J., Kerwin, J., Frechtling, J., & Vivari, B. A. (2002). Characteristics of the discussion in online an d face to face focus groups. Social Science Computer Review, 20 (1), 31 42. Schonlau, M., Fricker, R. D., & Elliott, M. N. (2002). Conduct ing research surveys via e mail a nd the Web. Santa Monica, CA: Rand Distribution Services. Retrieved August 24, 2010, f rom www.rand.org/pubs/monograph_reports/.../MR1480.ch3.pdf Schutz A (1967). The phenomenology of the social world Evanston, IL: Northwestern University Press Schwartz, J. A. (2008, May). Telehealth: Merging of t echnology and medicine leads to improved h e althcare. Progressive States Network Retrieved April 21, 2010 from www.progressivestates.org/content/840/telehealth merging of technology andmedicine leads to improved healthcare#4 Schwartz, J. R. (2008, September 30). Logging on for a second (or third) o pinion The New York Times p. F 2. Sc ott, J. (2000). Social network a nalysis (2 nd ed.). Thousand Oaks, CA: Sage Publications. Seidman, I. (1991). Interviewing as qualitative research: A guide for researchers in education and the social sciences. New York: Teachers College Press. Shah D. V. & Scheufele, D. A. ( 2006 ). Explicating opinion leadership: Nonpolitical dispositions, information consumption, and civic participation. Political Communication, 23 1 22 Shapiro, A. L., & Shapiro, R. C. (1999). The control revolution: How the in ternet is putting individuals in charge and changing the world we know New York: Public Affairs. Shapiro, D. L., Sheppard, B. H., & Cheraskin, L. ( 1992 ) Business on a handshake. Negotiation J ournal, 3(4) 365 377 Shih, T. H., & Fan, X. (2008). Comparing response rates from web and mail surveys: A meta analysis. Field Methods, 20 249 271. Shih, T. H., & Fan, X. (2008). Comparing response rates from web and mail surveys: A meta analysis. Field Methods, 20 249 271.

PAGE 242

242 Singh, A., Taneja, A., & Mangalaraj, G. (2009). Creating online surveys: Some wisdom from the trenches. IEEE Transactions on Professional Communication, 52 (2), 197 212. S innreich A., C hib A., & G ilbert J. (2008). Modeling information e quality: Social and media latency effects on information d iffusion International Journal of Communication, 2 132 159 Slater M D & Rounder D. (1996). How message evaluation and source attributes may influence credibility assessment and belief change. Journalism & Mass Communication Quarterly 73 974 9 91. Smith M. R., & Marx L. (E ds. ). (1994). Does technology drive history?: The dilemma of technological d eterminism Cambridge MA : MIT Press. Smith R. G (1978). The message measurement inve ntory: A profile for communication analysis. Bloomington IN: Indiana University Press. Spears, R., & Lea, M. (1994). Panacea or panopticon? The hidden power in computer mediated communication. Communication Research 21 427 459. Sproull, L., & Kiesl er, S (1991). Connections: New ways of working in the networked organization Cambridge, MA: The MIT Press. Stake, R. E. (1995). The art of case study Thousand Oaks, CA: Sage Publications. Stamme K & Dube R. (1994). The relationship of attitudinal comp onents to trust in media. Communication Research, 21 105 123 Strauss, A. L., & Corbin, J. M. (1998). Basics of qualitative research: Techniques and p rocedures for developing grounded theory. Thousand Oaks CA : Sage Publications. Streeter, C. L., & Gilles pie, D. F. (1992). Social network analysis. Journal of Social Service Research 16 201 222. Suitor, J., & Keeton, S. (1997). Once a friend, always a friend? Effects of homophily on Social Networks, 19 51 62. Tajf el, H. (1974). Social identity and intergroup behavior. Social Science Information 13 65 93. Tajfel, H. (1978). Differentiation between social groups: Studies in the social psychology of intergroup relations. London: Academic Press. Tajfel, H., & Turne r, J. C. (1981). The social identity theory of intergroup behavior. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (pp. 7 24). Chicago:Nelson Hall.

PAGE 243

243 Tanis, M. (2008). Health related on line f the big a ttraction? Journal of Health Communication, 13, 698 714. Taylor, S. J., & Bogdan, R. (1998 ). Introduction to qualitative research methods: A guidebook and resource (3 rd ed.). New York: John Wiley & Sons, Inc. Tilley, C. (1994). Space, place, landscape and perception: Phenom enological perspectives. In C. Tilley (Ed.), A phenomenology of landscape: Places, paths and monuments (pp. 7 34 ) Oxford: Berg. Turkle, S. (1995). Life on the screen: Identity in the age of the internet New York: Simon & Schuster. U.S. Department of Heal th and Human Services. (2009). Health, United States, 2009 With special feature on medical technology (Library of Congress Catalog Number 76 641496). Hyattsville, MD: U.S. Government Printing Office. Retrieved April 20, 2010, from http://www.cdc.gov/nchs/data/hus/hus09.pdf Valente, T. W. (1995). Network Models of the Diffusion of Innovations Cresskill, NJ: Hampton Press. Valente, T. W., & Pumpuang, P. (2007). Identifying opinion leaders to promo te behavior change. Health Education Behavior, 34 881 896 Walsh Childers, K. (1994). A death in the family: A case study of newspaper influence on health policy development. Journalism Quarterly, 71 (4), 820 829. Walther, J. B. (1992). Interpersonal effe cts in computer mediated interaction: A relational perspective. Communication Researc h, 19, 52 5 9 Walther, J. B. (1994). Anticipated ongoing interaction versus channel effects on relational communication in computer mediated interaction. H uman Communicati on Research, 20, 473 501. Walther, J. B. (1995) Relational aspects of co mputer mediated communication: E xperimental observations over time. Organization Science 6 186 203. Walther, J. B. (1996). Computer mediated communication: Impersonal, interpersonal and hyperpersonal interaction. Communication Research, 23 3 43 Walther, J. B., Anderson, J. F., & Park, D. W. (1994). Interpersonal effects of computer m ed iated c ommunication: A meta analysis of social and antisocial c ommunication. Communication Resear ch, 21 460 487 Walther, J. B., Gay, G., & Hancock, J. T. (2005). How do communication and tec hnology researchers study the internet ? Journal of Commun ication, 5 5 632 656.

PAGE 244

244 Wang, P. (1999). Methodologies and methods for user behavioral research. Annual Re view of Information Science and Technology, 34, 53 99. Wang Y. D., & Emurian H. H. (2005). An overview of online trust: Concepts, elements, and i mplications Computers in Human Behavior, 21 105 125 Wang, Z., Walther, J. B., Pingree S., & Hawkins, R.P. (2008). Health information, credibility, homophily, and influence via the internet: Web sites versus discussion g roups Health Communication 23 358 368 Warner, D., & Procaccino, J. D. (2007). Women seeking health information: Distinguishing the Web use r. Journal of Health Communication, 12 787 814. Warner W. L., & Lunt, P. S. ( 1941). The social life of a modern community New Haven, CT: Yale University Press. Wasko, M. M., & Faraj, S. (2000). It is what one does: W hy people participate and help others in electronic communities of practice. Journal of Strategic I nformation S ystems, 9 155 173. Wasserman, S., & Faust, K. (1994). Social network anal ysis: Methods and applications. Cambridge UK : Cambridge University Press. Watts, D. J. & Dodds, P. S. (20 07). Influentials, networks, and public o pinion f ormation Journal of Consumer Research, 34, 441 458 Watts, D. J., Dodds, P. S., & Newman, M. E. (2002). Identity and search in social networks. Science 296 ( 5571 ), 1302 1305. Weimann, G. (1991). The influ entials: Back to the concept of opinion leaders? Public Opinion Quarterly 55 267 279. Weimann, G. ( 1994 ). The influentials: People who influence people Albany NY : State University of New York Press. Weimann, G., Tustin, D., van Vuuren, D., & Joubert, J. (2007). Looking for opinion leaders: Traditional vs. modern measures in traditional societies. International Journal for Public Opinion Research 19 (2), 173 190. Wellman, B. (1983). Network analysis: Some basic principles. Sociological Theory, 1 155 2 00. Wellman, B. (1997). An electronic g roup is virtually a social network. I n S. Kiesler (Ed.), Culture of the internet (p p. 179 205 ). Mah wah, NJ: Lawrence Erlbaum Associates. Wellman, B., & Berkowitz, S. D. (Eds.). (1988). Introduction: Studying social s tructures. In S ocial structures: A network approach (pp. 1 13). New York : Cambridge University Press.

PAGE 245

245 Wellman, B., & Gulia M. ( 1999 ). Net surfers don t ride alone: Virtual communities as c ommunities. In M. Smith & P. Kollock (E ds. ), Communities in Cyb erspace (p p. 167 1 94 ) London: Routledge Wheeless, L. R. (1974) The effects of attitude, credibility and homophily on selective exposure to information. Speech Monographs, 41, 329 338. Williams, D. (2006). On and off the net: Scales for social capital in an online era (Electronic version). Journal of Computer Mediated Communication, 11 (2), 593 628. Retrieved March 27, 2010, from http://jcmc.indiana.edu/vol11/issue2/williams.html Willig C (2008). A phenomenological investigation of the experience of taking part in extreme sports. J ournal of Health Psychol ogy 13 (5) 690 702 Wilson S., & Peterson L. ( 2002 ). The anthropology of o nline c ommunities. Annual Review of Anthropology 31 449 4 67. Windahl, S. (1981). Uses and gratifications at the crossroads. Mass Communication Review Yearbook, 2 174 185. Wolff R. F. (1999). A phenomenological study of in church and televised w orship. Journal for the Scientific Study of Religion, 38 (2) 219 2 35 Wright, K. B. (1997). Shared ideology in Alcoholics Anonymous: A grounded theory approach. Journal of Hea lth Communication, 2, 83 99. Wright, K. B. (2000). Perce ptions of on line s upport providers: An e xamination of perceived homophily, s ource credibi lity, c ommunication and social support within o n line support g roups. Communication Quarterly, 48 ( 1 ), 44 59 Wright, K. B. (2004). On line relational maintenance strategies and perceptions of partners within exclusively internet based and primaril y interne t based relationships. Communication Studies, 55 239 253. Wright, K. B. (2005). Researching internet based populations: Advantages and disadvantages of online survey research, online questionnaire authoring software packages, and Web survey services (Elec tronic version). Journal of Computer Mediated Communication 10( 3) Retrieved April 11, 2010 from http://onlinelibrary.wiley.com.lp.hscl.ufl.edu/doi/10.1111/j.1083 6101.2005.tb00259.x/f Wright, K. B., & Bell, S. B. (2003). Health related support groups o n the internet: Linking empirical findings to social support and computer mediated communication theory. Journal of Health Psychology, 8 (1), 39 54.

PAGE 246

246 W u J. J., & Tsang, A. S. (2008). ehavior intention in virtu al communities Behaviour & Information Technology 27 (2), 115 125 Yin, R. (1994) Case study research ( 3 rd ed.). Newbury Park, CA: Sage Publications. Young, M., & Tseng, F. (2008). Interplay between physical and virtual settings for online interpersonal trust formation in knowledge sharing practice. CyberPsychology & Behavior 11 (1), 55 64. Zhang, Y. (2000). Using the internet for survey research: A case study Journal of the American Society for Information Science 51 (1), 57 68. Zhang, X., & Dong, D. (2008). Ways of identifying the opinion leaders in virtual communities (Electronic version). International Journal of Business and Management, 3 (7), 21 27. Retrieved February 22, 2010, from http://ccsenet.org/journal/index.php/ijbm/article/viewFile/1392/1354 Zimmer, M. (2008). Critical perspectives on Web 2.0. First Monday, 13 (3). Retrieved March 16, 20 1 1, from http://www.uic.edu/htbin/cgiwrap/bin/ojs/index.php/fm/issue/v iew/263

PAGE 247

247 BIOGRAPHICAL SKETCH Michael D. Brookshire was born in Hollywood, California, and came to higher education later in life. He was a carpenter for a number of years before graduating from Fullerton College in Fullerton, California, with an Asso ciate in Arts Degree in 1988. He film in 1990. Upon graduation, he returned to the Los Angeles area and began a f ree lance career as a writer, videographer and video editor for various corporate, industrial, informational, training and sales video productions Clients included Newport Beach (CA) Fire Department, University of La Verne ( CA), Security Pacific Bank, North Orange County (CA) Community College District, Garden Grove (CA) Unified School District. In 1994, Mr. Brookshire took his video production knowledge and expertise to Burlingame Industries, Inc., a diversified manufacturi ng and sales company in Rialto, California, where he became Video Production Manager, a position he continues to hold today. Mr. Brookshire wrote, produced and directed numerous video programs for the company while earning his Masters Degree in communicati ons at California State University, Fullerton and teaching television studio production at Fullerton College. In 2005 he negotiated a part time position at the company in order to pursue a doctorate in mass communications at the University of Florida. As a graduate teaching assistant at Florida, Mr. Brookshire taught entry level, intermediate, and advanced television studio and video production courses from Fall 2005 through Fall 2010, and received his doctorate in Spring 2011. He is a p ermanent member of Alpha Gamma Sigma at Fullerton College, was a T York University from 1988 through 1990, won an A ward of Excellence from The

PAGE 248

248 Videographer Awards in 1999, was a Giles T. Brown Award nominee for outstanding thesis at Cal ifornia State Un iversity, Fullerton in 2005, and received a G rinter Fellowship from the College of Journalism and Communications at the Universi ty of Florida in 2009.