Informational support exchanges using different computer‐mediated ...

Informational Support Exchanges Using Different Computer-Mediated Communication Formats in a Social Media Alcoholism Community

Katherine Y. Chuang and Christopher C. Yang College of Computing and Informatics, Drexel University, Philadelphia, PA 19104. E-mail: {katychuang, chris.yang}@drexel.edu

E-patients seeking information online often seek specific advice related to coping with their health condition(s) among social networking sites. They may be looking for social connectivity with compassionate strangers who may have experienced similar situations to share opinions and experiences rather than for authoritative medical information. Previous studies document distinct technological features and different levels of social support interaction patterns. It is expected that the design of the social media functions will have an impact on the user behavior of social support exchange. In this part of a multipart study, we investigate the social support types, in particular information support types, across multiple computer-mediated communication formats (forum, journal, and notes) within an alcoholism community using descriptive content analysis on 3 months of data from a MedHelp online peer support community. We present the results of identified informational support types including advice, referral, fact, personal experiences, and opinions, either offered or requested. Fact type was exchanged most often among the messages; however, there were some different patterns between notes and journal posts. Notes were used for maintaining relationships rather than as a main source for seeking information. Notes were similar to comments made to journal posts, which may indicate the friendship between journal readers and the author. These findings suggest that users may have initially joined the MedHelp Alcoholism Community for information-seeking purposes but continue participation even after they have completed with information gathering because of the relationships they formed with community members through social media features.

Introduction The Internet is often used for social reasons (Høybye, Johansen, & Tjørnhøj-Thomsen, 2005; Tufekci, 2008) and also for personal reasons, such as seeking health informaReceived September 26, 2012; revised February 15, 2013; accepted February 18, 2013 © 2013 ASIS&T • Published online 23 October 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/asi.22960

tion, because of the ease in reaching a broader group of users combined with the option of anonymity if desired. In recent years, Internet users have been increasingly sharing their health information in online social networks, gathering information and seeking support as they face important healthcare decisions (Boase, Rainie, Horrigan, & Wellman, 2006; Fox & Jones, 2009). Typically, e-patients want to access user-generated or “just-in-time someone-like-me” health information community-type websites such as newsgroups, blogs, social networking sites (SNS), or status updates (Fox & Jones, 2009). Those e-patients that seek, share, and sometimes create information about health and wellness benefit from sharing their experiences, discussing medical information, and exchanging social support (Fox & Jones, 2009). E-patients who participate in exchanging social support in these communities receive additional benefits such as improved ability to cope with stress and increased access to relevant information (Cohen, 2004; McCormack, 2010). Previous research literature suggests that online communities become surrogate families of e-patients, in which members share common problems, help each other toward mutual goals, and support each other through good times and bad (Radin, 2001; Wright & Bell, 2003). Many users join online support groups for a sense of community with those who have experienced similar situations because they are more likely to have compassion and related experiential knowledge (Radin, 2000; Wright & Bell, 2003). Online communities are enabled through social software technologies (i.e., e-mail lists, discussion boards) and most recently through websites known as social networking sites SNS. The SNS features enable users to find each other and build connections using profile pages and a spectrum of private and public communication tools (Ahn, Han, Kwak, Moon, & Jeong, 2007; boyd & Ellison, 2007). SNS is unique from previous text-based communication formats because of its emphasis on representing relationships between users and “architectural elements” that encourage interpersonal relationships (Papacharissi, 2009). In this

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY, 65(1):37–52, 2014

article, architectural elements refer to website design features that designate writing privileges and features that allow for modifying access to a user’s written messages. Studying the interactions within a support community will give useful insight into the users’ information needs from the group and their behavioral patterns. It could contribute to improving the design of online intervention programs by suggesting new software features to promote a support environment. The objective of this research is to identify the types of informational social support across the following three different text-based computer-mediated communication (CMC) tools: discussion forums, personal journals, and notes. This study presents empirical data as evidence demonstrating that patterns of informational support vary among the different online tools. Results are then discussed in conjunction with theories of waves of information needs during the course of health information seeking (Radin, 2000) and the blend of public and private space (Donath, 2007; Meyrowitz, 1986) to explain possible causes of these patterns. Prior Work Much of the previous related research demonstrates that e-patients’ information needs not fully met by medical information offered in the clinic (Wright & Bell, 2003). In addition, e-patients seek assistance beyond technical support, such as reminders to visit a doctor and to take medication, relating to managing their healthcare. Although previously documented research studies examined the aspect of social support exchanged online, their operational definitions for online communities focused on the expressions of social support rather than distinct separation of conversations from the communication platform. We bridge this gap by studying the impact of CMC format on the interpersonal communication behavior of individuals participating in a public online support group. Users often conduct their conversations online as a private space even though it is quite public (Papacharissi, 2009; Preece, 2000; Smith, 2004; Walstrom, 2004). In this section, we describe social support for alcoholics, architectural elements of CMC, and the MedHelp Alcoholism Community. Social support through online support groups. Consumer health information needs are met through a variety of sources, including conversations within peer support communities (Civan & Pratt, 2007; Klaw, Heubsch, & Humphreys, 2000; Preece, 1999; Wright & Bell, 2003). These mutual help groups offer different types of social support (informational, nurturant, and instrumental). These health-related social media sites are good sources for people, such as alcoholics, to gather information anonymously to avoid the stigma that comes from traditional face-to-face (F2F) interactions. Support community members often share experiences, which may include personal information. Information types found in these messages can vary from treatment options, 38

links to other websites, or book recommendations. Exchanging emotional support is important for building a supportive environment. In addition, peer communication, such as establishing social norms or finding role models and sharing feelings, can also play a role in facilitating new health habits, such as quitting smoking (Ancker et al., 2009). However, there is little documentation about the information types present in journals or profile posts. Informational support in online support groups. Previous research studying online support communities for various health conditions found that the participants exchange different types of information and also discuss benefits for e-patients in these groups; these informational support types (advice, teaching, fact, opinion, information referral, and personal story) are discussed more in the coding scheme section (Bambina, 2007; Braithwaite, Waldron, & Finn, 1999; Civan & Pratt, 2007; Coursaris & Liu, 2009; Cunningham, van Mierlo, & Fournier, 2008; Eichhorn, 2008; McCormack, 2010; Pfeil & Zaphiris, 2007; Wright & Bell, 2003). Although a range of health conditions is studied, each paper reports findings of only one communication format (i.e., forums), instead of multiple communication formats for the same user base. This shows a lack of analysis that studies the relationship between the use of the tool and the tool available to the same community members. In this study, we compare different CMC formats through identification of the effect that the architectural elements of each CMC format has on supportive interaction patterns. Types of information on alcoholism online. Alcoholism is a form of dependence, characterized by compulsive and uncontrollable alcohol consumption despite its negative effect s on the drinker’s health, relationships, and social standing (National Institutes of Health1). Alcohol abuse is a serious problem and is on the list of public health issues (Centers for Disease Control and Prevention, World Health Organization2). Every year, harmful use of alcohol is accountable for at least 2.5 million deaths (World Health Organization, 2010). Alcohol abuse is a habit that commonly originates in groups, and therefore alcoholics should use social methods such as support groups for interventions that can result in long-term behavior change (Rosenquist et al., 2010). Treatment often consists of participation in support groups (Alcoholics Anonymous, 2001; Beattie & Longabaugh, 1999; Rosenquist et al., 2010) or in recent cases, online intervention (Cunningham et al., 2006; Linke et al., 2007). Most health information online regarding alcoholism can be divided into the pro-drinking or anti-drinking camps

1

Department of Health and Human Services, Alcohol Dependence (Alcoholism) (PDF), National Institutes of Health. 2 Statistics of alcohol abuse: http://www.cdc.gov/nchs/fastats/ alcohol.htm, http://www.who.int/substance_abuse/facts/alcohol/en/index .html

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

(Joinson & Banyard, 2003). Online support community members can speak on behalf of both sides and offer personal experiences to support either argument. These insights are part of the social aspect that draws people to join and participate in the community. Because of the stigma surrounding alcoholism, it is more convenient to seek alcohol-related information online. Available information includes either pro-drinking or anti-drinking articles from reputable sources (Joinson & Banyard, 2003). E-patients interested in alcohol information can also search for social information such as a Question and Answer (Q&A) website or online support groups. Computer-mediated communication through social networking sites. In SNS, users must balance their private and public selves, especially when conducting social interactions with each other (Meyrowitz, 1987; Papacharissi, 2009). For example, in a physical space, the walls of offices and houses create distinct spaces. However, this distinction is not as clear on the Internet. Public and private information is blurred in the online space because the architectural features of a website are more transparent, making it difficult to separate distinct spaces or develop behavioral norms (Papacharissi, 2009). Online communication encourages disparate individuals to connect, communicate, and take action, especially through SNS, which fosters interaction that is primarily interpersonal (Ellison et al., 2009; Papacharissi, 2009) Identity information from online profiles can assist in developing common ground and facilitate initial interactions (Ellison et al., 2009). SNS interactions are everyday face-to-face interactions adapted to an online setting, which because of the text-based asynchronous conversation, will blur the distinction between public and private spheres into a combined space. Smith (2004) described this lack of distinction as confusing space that is publicly accessible but used in a private way. For this study, we take this view of publicly available, user-created messages as those used in a personal way for interpersonal communication (private arena) and mass communication (public arena). When analyzing the social setting of computer-mediated communication, social support exchanges can be thought of in economic terms. In a successful supportive interaction exchange, there will be a combination of (a) someone offering and (b) someone requesting support. People offer support because of altruistic reasons (McLure-Wasko & Faraj, 2000). People may ask for help when there is an information gap, and altruism may be the result of an internal desire to help others. On the other hand, when considering support exchanges from an architectural perspective, we see a relationship between the site design and user behavior (Donath, 2007; Meyrowitz, 1987; Papacharissi, 2009). Site design “promotes the development of particular culture or behaviors and identity presentation,” which may be found through studying interactions rather than explicit articulations on profile pages (Donath, 2007, p. 236). For instance, public comments and other communication can

signal the “strength and context of a relationship” (Donath, 2007, p. 238). An example of a public space is an urban space, which has characteristics that allow public gatherings to happen in a neutral space, similar to an online forum that provides a grassroots setting (Oldenburg, 1989; Putnam, 1995). The space experience comes from relations with others (Humphreys, 2007; Lofland, 2006). In these neutral spaces, the core group’s main activity is informal conversation. Public spaces such as parks are communal and have certain purposes of use when compared with a private space such as a home. Communication tools are designed with a spectrum of privacy options. SNS provide the capability for users to represent themselves with an online presence that contains shareable personal information, such as birthdays, preferences, photographs, writings, and so on (Boyd & Ellison, 2007). Convenient features allow users to form and maintain online network “friends,” where if one user invites another user to be friend and is accepted, a relationship is established on the website (Ahn et al., 2007). Friends can communicate through SNS in several ways, including private and public messaging (Thelwall & Wilkinson, 2010). What we are studying in the communication within these community spaces, structured by site design, are what Donath (2007) refers to as signals of social information and behavior that can indicate an individual’s social status. For example, returning to a friend’s page with a greeting of hello is a way of indicating a relationship, whereas mentioning mutual friends and external events are references of shared experience. The length, frequency, and content of these comments form a conversational profile of each relationship (Donath, 2007; Papacharissi, 2009). As Papacharissi (2009) states, online social networks suggest acceptable behavior through their architectural elements, and in turn the users customize them to improve their social interactions. “Facebook emerged as the architectural equivalent of a glass house, with a publicly open structure which may be manipulated (relatively, at this point) from within to create more or less private spaces.” (Papacharissi, 2009, p. 199). The loose behavioral norms suggested by site design encourage users to construct their own behavioral norms using the tools provided to leave cues for each other. They can also use the tools for self-presentation. In a more tightly administered setting such as LinkedIn, individuals are restricted in terms of self-presentation and thus have less spontaneous interaction: “the spaces created are relatively more private and open to only elite and professional publics.” (Papacharissi, 2009, p. 215). Public and private spheres are spatially communicated at three levels: situating the architectural foundation, development of sociocultural categories, and the controls for setting privacy; these lead to the development of either loose or tight behavioral norms (Papacharissi, 2009). MedHelp Support Communities MedHelp is a health-oriented SNS helping individuals connect with people and to information resources.

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

39

FIG. 1. Screenshot of forum messages listed on profile page. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

FIG. 2. Screenshot of journal messages listed on profile page. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

It provides a platform for peer support communities where any registered user can become a member and access several interpersonal communication tools, such as discussion forums, journals, and notes. The forum for each peer support community is unmoderated, where any MedHelp user is able to post questions or respond with comments (Figure 1). Journals allow users to record thoughts and feelings (Figure 2). Notes are a way for users to keep in touch with each other on MedHelp (Figure 3). Although the messages on each of these CMC formats may be displayed publicly depending on individual privacy settings (public, friends, and private) controlled by the user, the literature review suggests that user behavior norms are more likely to be swayed by overall ability to provide individual taste profiles, such as favorite music and brands, than the transparency of any content (Papacharissi, 2009). Architectural elements for each of the MedHelp SNS CMC formats are listed in Table 1. When posting to the forum and journals, ticker data such as daily weight (Figure 2) can be added. All forum content created by users is set to be publicly accessible. However, MedHelp allows journals and notes to be set to one of three options: “Everyone,” “Only my friends,” and “Only me.” New posts to the forum can be viewed on the forum page, which is also known as the support community page. Updates to public journals (new

TABLE 1.

40

Comparison of CMC architectural elements.

CMC

Access

Tags

Other features Select topic, Add to watch list, Show Ticker Show Ticker, Add Photo Add as friend

Forum

Public

Yes

Journal

Everyone, Only my friends, Only Me Everyone, Only my friends, Only Me

Yes

Notes

posts or new comments) are listed on the support community page under the “recent activity” box. There is also a section that lists community members with links to their profile pages. Each profile page displays sections of the user’s activity on the communication tools. Privacy settings may affect what can be viewed on a user’s profile page. If the setting for journal and notes are set to only “my friends” then only users who are “friended” may view this content. If the setting is to “only me”, only the user can see his or her own content when logged in. The content in each of these tools is organized chronologically. Architectural elements in MedHelp. The three communication tools we investigate in this study are forum, journals, and notes. They are available for any MedHelp support

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

FIG. 4. Posting forum question. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

FIG. 5. Posting journal post. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

FIG. 3. Screenshot of note messages listed on profile page. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

community member to post content. Each tool has variation in features for posted content (Figures 4 to 7). Users can post questions or polls to the forum. They are required to fill out a title, select a topic, and describe their question and are free to add tags (Figure 4). Posting to journals can include title, entry, tags, and photos, with selected privacy options (Figure 5). Posting notes on a user’s profile includes type of note and the content in the note (Figure 6). If the user is not

a friend, there is an option to befriend the user. The screen shots are shown in Figures 1–3. In this study, we explore some possible reasons why people are drawn to online health supportive communities by looking at supporting interactions. We hypothesize that patterns of informational support exchanged across different social media communication formats vary, depending on architectural elements (i.e., access, ownership). In this study, we try to answer the following question: What are the different levels of informational support across different communication formats within the same community? In the discussion, we attempt to explain why these patterns are different. Method Our approach consisted of collecting data with a web crawler, formatting the data, and conducting content analysis.

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

41

the data—informational support and nurturant support. The third type, instrumental support, is typically found in faceto-face support interactions and not found in the data for this study. Support Types—Informational Versus Nurturant Support FIG. 6. Posting note. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

FIG. 7. Ticker data. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Data Collection and Preparation We downloaded data from MedHelp Alcoholism Community (www.medhelp.com) using a web crawler during September 2009 and selected a 3-month period (June 19, 2009 to September 19, 2009). Data include publicly viewable messages from three CMC formats created by users belonging to the community: discussion forums, nF = 493; user journals, nJ = 423; and posts on users’ profile pages called “notes,” nN = 1,180. The messages in each sample were converted to spreadsheets for content analysis, with each individual message as the unit of analysis. Content Analysis We used descriptive content analysis to identify social support exchange patterns in the messages of the MedHelp Alcoholism Community. Messages were the unit of analysis, with the possibility of multiple codes appearing. Definitions of social support types were developed by reviewing examples from related literature as well as matching them with themes presented within our data (Bambina, 2007; Cutrona & Suhr, 1992; McKenna, Green, & Gleason, 2002; Winzelberg, 1997). Concepts and their definitions were drawn out and organized into three main categories (informational, nurturant, instrumental), suggested by Cutrona and Suhr (1992). Only two types of support were found in 42

Social support is the provision of psychological and tangible resources intended to benefit an individual’s ability to cope with stress, such as information leading the subject to believe that he or she is cared for. Researchers describe two primary themes of social support that are exchanged in a supportive relationship: (a) action facilitation, which is intended to assist a stressed individual to solve or eliminate problems causing distress, and (b) nurturant support, which caters to the emotional side by comforting support seekers (Cutrona & Suhr, 1992). Action facilitation support includes both informational support and instrumental support. Informational support could be facts, advice, information referral, personal stories, or opinions. Instrumental supports are tangible services, either direct or indirect, that can improve a situation, (e.g., driving someone to the hospital). In this portion of a multipart study, we focus on the informational support that occurs in different CMC formats of an online health support community. Although an online support group offers both informational and nurturant support, here we are interested in comparing the CMC formats with regard to informational needs. We look for patterns of informational support type interactions across each format to identify differences at the group level, namely advice, fact, opinion, information referral, and personal stories, which are defined in more detail in the next section. Coding scheme—informational support types. Information types identified from the data set include advice, fact, opinion, information referral, or personal story. We used the social support categories from Cutrona and Suhr’s (1992) study as a base for a final coding scheme. First, the codes were compared against messages in the forum to verify that all five types exist. Next, we adjusted the definitions by comparing our definitions and examples with similar studies that also used content analysis to analyze online support group messages. These studies provided information to help our process of creating a coding scheme. To protect against being overly subjective when assigning definitions to the social support types, we selected studies that cover a broad range of health conditions spanning a decade of research rather than just the past few years (Tables 2 and 3). The studies that publish the number of participants, date range, and number of messages reviewed ranged from a small group (42 users in 1999) of approximately 1,500 messages to 5,000 messages in 2009. By examining the informational support types identified in those studies, we found that they ranged from physically limiting conditions (i.e., disability, cancer, anterior cruciate ligament [ACL] injuries) to mental health issues (i.e., alcoholism, depression) and contained the same five subtypes.

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

TABLE 2.

Related studies.

Code

Year

B99

1999

K00 B07

2000 2007

Braithwaite et al. Klaw et al. Bambina

C07

2007

Civan & Pratt

P07

2007

C08

2008

E08

2008

Pfiel & Zaphiris Cunningham et al. Eichhorn

C09

2009

M10

2010

TABLE 3.

Author

Most of the previous studies looked at specific types of informational support (Bambina, 2007; Braithwaite et al., 1999; Civan & Pratt, 2007; Coursaris & Liu, 2009; Cunningham et al., 2008; Klaw et al., 2000; McCormack, 2010; Pfiel & Zaphiris, 2007). However, two of the studies were broader, mentioning informational support without going into detail (Eichhorn, 2008; Preece, 1999). Table 4 summarizes the data, CMC format, and results of other studies.

Data

Coursaris & Liu McCormack

“Support Network” (disabilities), 42 users, 1,472 messages Alcoholics mutual help OnLine Cancer Forum, 84 members, 1,149 messages 3 Breast cancer forums 163 users, 458 messages 78 users, 246 messages 77 users, 312 messages Depression/seniors, 47 users, 7 months, 400 messages Alcoholism, 50 users, 10 months, 474 messages (moderated group) 5 eating disorder message boards, 490 randomly selected messages 5 HIV/AIDS groups, 5,000 randomly selected messages Eating disorder, 95 users, 17 days, 325 messages

Intercoder Reliability Once the coding scheme was finalized, the first coder went through the data samples to code for the five informational support types. The second coder performed content analysis with the same coding scheme. Two independent coders separately coded the messages with informational support types to validate the coding scheme. Cohen’s kappa was calculated at 0.734, which shows that the code used for this study is reliable.

Comparison of informational support types. Studies & settings

Support type

Disability

(All) Advice/teaching Fact (situation appraisal) Opinion Information referral Personal story

TABLE 4. Year

Eating Disorder

HIV/ Aids

Alcohol

C09 C09

C07 C07 / K00 C07 K00 C07 / K00

E08 M10

B99 B99

C09 M10

Depression/ Psychosis

Cancer

P07 / C09b P07 / C09b

C07 C07 C07 C07 C07

P99

M10 B99

Anterior Cruciate Ligament (ACL)

P07

Summary of the related literature. Author

Data

1999

Preece

Torn Knee Ligament, 500 msgs, April 1996 – April 1997 “Support Network” 42 users, 1,472 messages

1999

Braithwaite et al.

2000 2007

Klaw et al. Bambina

Online alcoholism support group Support OnLine Cancer Forum 84 members 1,149 messages (unmoderated)

2007

Civan & Pratt

Breast cancer

2007 2008

Pfiel & Zaphiris Cunningham et al.

2008

Eichhorn

2009

Coursaris & Liu

2010

McCormack

Depression/seniors Alcoholism 10 months; 474 posts (moderated) 5 eating disorder msg boards 490 messages HIV/AIDS 5,000 messages Anorexia bulletin board

Architectural elements E-mail list “Messages were distributed via E-mail through a nationwide computer BBS network.” Mailing list “an Internet cancer support forum . . . a virtual space . . . [to exchange] support” Only requires e-mail address. Archives posted online and publically available Bulletin boards

Patterns Nurturant > Information (no distinctions) P>F 31.3% Informational

Self-disclosure > Information/Advice 38.6% Informational

Bulletin board Bulletin board

Board A 68.3% info Board B 40% info Board C 70% info A>R P>A>F P>A>F>O

Yahoo discussion groups

29.7% Informational

Bulletin board

41.6% Informational

Bulletin board

A=F=R>P>O

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

43

Definitions used in content analysis. Information support describes messages that convey instructions, including (a) advice or teaching, (b) referrals to other sources of information, (c) facts, (d) stories of personal experience, and (e) opinions. Messages coded as information support often appeared as an attempt to reduce uncertainty for the message recipient (Braithwaite et al., 1999). Advice (also referred to as teaching) offers ideas and suggests actions for coping with challenges. For example, it provides detailed information facts or news about the situation or skills needed to deal with the situation. A member asks for suggestions on quitting drinking behavior in the forum by mentioning a personal story that led to alcohol consumption, “I had not drunk even a drop of beer till the age of 30 YRS , but after marriage , I started taking alcohol and became a daily consumer with drinks QTY upto 500 Ml many times.” The message continues with his information need from the forum, “Now, I want to quit, but am not able to do so as my wife always gives me tensions,” while addressing the need for advice to his specific problem: “What should I do?” Members of the community also asked for advice and support regarding relationships they had outside their electronic circle of friends. For example, one community member asked forum members for advice about dealing with dysfunctional family members, “What should I be doing? Should a person just not think about their dysfunctional family? Is that how you detach?” Another member asks for advice on side effects related to alcohol consumption by introducing the problem: “My husband has a sickly sweet body odor after drinking even one drink.” She conveys her problem with specific examples, “[I] can tell he’s had even one beer after about 30 minutes if he comes within 2 feet of me. on nights he drinks a lot, the entire room takes on the sweet smell” and then goes into the questions asking if professional consultation with a doctor is necessary because it is something she has never encountered before “half the people in my family are drinkers, and I’ve never smelt this until I met him.” Referral to information occurs when the recipient asks for information sources, or it may refer to efforts to link the recipient with a source of expertise. For example, this person wrote in the forum about an author, “The book [by Dr. Edward Hallowell] I mentioned may be out of print but he has written many on anxiety.” and also wrote an opinion on the quality of the book “Driven to Distraction is a good one!” It may also be a request to find out how to get access to the information: “How can I get to the video link? Thanks so much.” Facts entail situation appraisal, when someone reassesses the situation, often exhibited when someone provide a different way to look at things. For example, one forum comment explicates on Post Acute Withdrawal Syndrome (PAWS): “[PAWS] comes from years of heavy drinking” by specifying, “[It] takes a LONG time for the central nervous system to repair itself”, yet also offering support, 44

“I think [you] get impatient with this [repair process,] understandably so” to show compassion. Factual information could also describe situations or present technical information. For example, one user describes a recent trip by a local bus in a journal entry, “It was an all day trip but [the] first time I went anywhere by myself since out of the hospital. I now have a little more freedom and independence.” Ending the journal post with some encouragement, “God continues to answer prayers.” Users also write about their daily habits in the journal format, for example, one user wrote a journal post with the text, “dry as a bone today” with the Addiction Recovery Tracker tool. Although we cannot make assumptions about why this message was written because it was not attached to a question, in our analysis this kind of message is considered offered information. Personal experiences are messages about a person’s experiences or incidents. It has a more story-like form that is about self-disclosure and possibly personal information. For instance, one forum respondent commented on a post saying “I have 25 years sober/clean” also offering opinion and encouragement, “Every day u don’t drink or use its a sober clean day! I think you x-wife is so very wrong! Nobody passed and appointed her an AA goddess!).” Another user starts a thread in the forum with a personal experience to ask for advice, “I was drinking every night or every other night for over a year. I would drink 6 to 8 beers at night.” With specific details, “Never got wasted [for] over 6 months and I don’t feel like I want to drink but I remember whenever I had a few drinks at least I could laugh and enjoy watching tv and minor things like that” ending with request for factual information, “Do I have depression? Did alcohol made permanent damage after drinking a 6 pack a day for over a year?” Opinions are a form of feedback, which can be a view or judgment formed about something. It is not necessarily based on fact or knowledge. One forum post asks for an opinion by starting with some opinions, “I like to have a vodka martini followed by 1 beer every nite” in addition to some personal situations, “On [the weekend, I’ll] have about 6 or 7 drinks (out to dinner or a show with wife)” to end with feedback “What do you all think? Do I have a problem?” In the comments, users provide opinions for all sorts of things, including reviews of the Atanabuse medication: “[It] has two side effects men hate . . . onion breath and erectile dysfunction.”

Results The three samples were user-created messages from the discussion forums (nF = 493), the user journals (nJ = 423), and the profile posts (nN = 1,180). The messages in forums and journals format were grouped into posts (i.e., messages that start the thread) and comments to the post, which ended

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

up in four data samples and the fifth being notes. There were 81 forum posts (81) and 412 forum comments (FC). There were 88 journal posts (JP) and 335 journal comments (JC). We first identified informational and nurturant support in the samples (both provided and requested) and present these findings first before presenting specific informational support types. One message contains 2.57 codes on average with a maximum of 10 codes per message, except for the first post of each thread, which had maximum of six codes. Some messages only offer support (i.e., “Have you tried Naltrexone? It is supposed [sic] to help with the cravings there are other meds that can help with it too. If all else fails, make a picture of tea and pop some popcorn and hang out with him with your “drink”), or only request support (i.e., “Hi, is there a medicine to take to stop the craving for alcoholic drink?”). Informational Versus Nurturant Support Our initial coding of the broad level of informational and nurturant support shows a contrast in results (Figures 8 and 9). Discussion forums were more informational than journals and notes based on requested information types. Additionally, more nurturant support is offered in notes and journal comments, whereas more information support is offered in the forum. When requesting social support, users of this community were not as likely to seek nurturant support, except in the journal posts. This imbalance in which nurturant support is high in journals and notes and low in forums persuades us to investigate more carefully the subtypes of informational support to better understand the link between supportive interactions and communication across different CMC formats.

FIG. 8.

Although an online support group offers a broad range of social support, here we are comparing the CMC formats with regard to informational needs. Informational Support Types We looked for the five different types of information across all the messages. Across all CMC format messages, fact was the most exchanged type of information in comparison with the other support types. Notes and JC showed similar patterns of behaviors for both offered and requested. JP, FP, and FC showed similar patterns for requested informational support. The style of writing in notes was typically different from forum and journal formats when requesting information. In forum messages and journal posts, users were likely to request information types other than fact; however, in notes users are more likely to exchange facts more directly without easing into it with stories or information referrals. This is different in the longer messages of journals and forums, which contain more stories, opinions, and advice. There was a loose linkage between offered and requested support, for example, advice is offered in the forum comments but requested in forum posts. Advice is offered in the journal comments but not requested in journal posts. For all the samples, fact is exchanged in most. Fact occurred most in notes messages (74.0% offered; 96.7% requested), whereas the other information types are less frequent. Advice is offered (13.6%) but not requested as much (0.4%). Similarly, opinion is offered (7.2%) more than requested (2.1%). This could be an indication of using notes format for altruistic reasons. Second, in journal messages, the high

Support types offered.

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

45

FIG. 9.

FIG. 10.

46

Support types requested.

Information support types offered. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

FIG. 11.

Information support types requested. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

incidence of fact offered in JP (60.2%) suggests that users were documenting their thoughts. Among JP messages there was a combination of referral (19.4%), personal stories (8.7%), and opinions (8.7%), and a bit of advice (2.9%) offered; yet for requested support there is less for the categories fact (52.9%), opinion (17.6%), and advice (29.4%). This may be because they did not expect responses, unlike JC messages where fact is very high (96.4%), and some advice is requested (3.6%). JC messages were most likely to offer fact (66.0%) along with some advice (21.9%). JP and JC matched in the exchange of advice and opinion. JP requests opinion and advice in addition to fact, and JC offers these three types more than the other informational types. More specifically, JC had high levels of fact for both offered (66.0%) and requested (96.4%); advice offered less (21.9%) and requested (3.6%). Opinion, personal stories, and referrals were not requested at all, and offered in few messages (O: 7.0%, P: 2.8%, R: 2.3%). In JP, fact was exchanged in most messages (60.2% offered; 52.9% requested). Personal experience and referrals were not requested at all (0%), but referral was offered in some messages (19.4%). This pattern of primarily fact providing and requesting suggests that journals may be a place for sharing

information. Also, some personal experience was offered in JP (8.7%). Other informational support types were requested and given. Advice was sought in more than a quarter of the messages (29.4%) but not offered as much. JP may also be a good place to seek advice, as comments contain advice. The high incidence of advice and opinion exchanges in the journal format suggests that the users may have a strong relationship, such as being close friends who are comfortable sharing their thoughts. Although forum posts and forum comments also had a relationship, it followed a slightly different pattern. FP messages offered personal stories (31.0%), facts (69.0%), or a combination of these two. FP requested mostly facts (49.4%), advice (27.8%), and opinion (16.5%) and returned in FC (O: 10.1%; P: 13.9%; a: 24.2%), but also more personal stories than expected (1.3% requested in posts but 10.1% offered in comments). In more detail, we see that the relationship between FP and FC messages is a polite and altruistic exchange, where more is given than requested. In FC format, the theme of personal experience was provided in a fraction of the messages (13.9%), but fewer messages requested this type of support (2.9%). Opinion (10.1%) and advice (24.2%) were also similarly offered in messages. Conversely, for

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

47

TABLE 5.

Summary of samples.

TABLE 7.

Sample

Forums (129 users) Journals (134 users) Notes (356 users)

Type

Size 493 messages Posts FP = 81 messages Comments FC = 412 messages

TABLE 6.

423 messages JP = 88 messages JC = 335 messages

Rank #1 #2 #3 #4 #5

Journals

Notes

FP

FC

JP

JC

N

0 0 60 27 0 87

134 20 267 77 56 554

3 20 62 9 9 103

47 5 142 6 15 215

110 25 598 17 58 808

requested support, there was a similar number (10.4%) and a different fraction for advice (11.8%). Referral was not requested at all but was given in some messages (13.9%). Advice was requested in nearly a third of the FP messages (27.8%) and opinions in 16.5%, but neither type was offered (0%). When starting forum threads, many messages shared facts (69.0%) or personal experiences (31.0%). This may be a strategy to obtain advice, stories, and opinions because offering opinions may not be helpful in seeking advice from others. JP has the highest percentage of messages with referrals, even though forums seem to have more information flow. This may be because users are recording information they discover.

Opinion

Personal

Referral

Advice

N 74.0% FP 69.0% JC 66.0% JP 60.2% FC 48.2%

FC 10.1% JP 8.7% N 7.2% JC 7.0% FP 0%

FP 31.0% FC 13.9% JP 8.7% JC 2.8% N 2.1%

JP 19.4% FC 3.6% N 3.1% JC 2.3% FP 0%

FC 24.2% JC 21.9% N 13.6% JP 2.9% FP 0%

There are also patterns that can be found among the samples when studying in depth the kinds of support that were offered. Table 5 summarizes the number of messages that indicate informational support offered by subtype; Figure 10 shows this as a bar graph. Table 6 shows a distribution of the support types and ranks by information type. Forum comments were more likely to have messages offering opinion (10.1%) than journals or notes. Forum posts were most likely to offer personal stories (31.0%). Across the samples, we observed a pattern where the information types in the middle are opinion and personal stories. Fact is always the highest, and advice and referrals interchange positions: JC similar to Notes JP opposing FC FC similar to JC

TABLE 8.

Number of messages requesting informational support. Forums

Advice Referral Fact Personal Opinion Total

TABLE 9. Type Rank #1 #2 #3 #4 #5

Journals

Notes

FP

FC

JP

JC

N

22 4 39 1 13 79

8 0 51 2 7 68

5 0 9 0 3 17

1 0 27 0 0 28

1 1 231 1 5 239

Ranking of samples by information type requested. Fact

Opinion

Personal

Referral

Advice

N 96.7% JC 96.4% FC 75.0% JP 52.9% FP 49.4%

JP 17.6% FP 16.5% FC 10.3% N 2.1% JC 0%

FC 2.9% FP 1.3% N 0.4% JC 0% JP 0%

FP 5.1% N 0.4% JC 0% JP 0% FC 0%

JP 29.4% FP 27.8% FC 11.8% JC 3.6% N 0.4%

highest, very likely because the users joined the online community for information but stay because of the community.

Informational Support Offered

A > O > P/R R > O/P > A vs. A > O/P > R A > O/P/R

The only one not following these patterns was the FP message that offered fact and personal stories only. Fact is occurring in all the message formats and it is also the 48

Fact

Number of messages offering informational support. Forums

Advice Referral Fact Personal Opinion Total

N = 1,180 messages

Ranking of samples by information type offered.

Informational Support Requested The number of messages identified with requested informational support was less than the number of messages that offered informational support. We calculated percentage by dividing by total number of messages in the sample by the number of messages that contained informational support and then displayed these percentages in a bar graph (Figure 11). Tables 7 and 8 shows a distribution of the support types and ranks by information type. Table 9 shows the ranking of CMC messages by fact type. Comparing across samples, all CMC formats request advice quite heavily except notes. FP, FC, JP Notes & JC

F>A>O F

Perhaps notes were not the main source for information seeking but, rather, for maintaining relationships. The pattern found among FP, FC, and JP is that there is a similar

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

combination of advice greater than opinion. There are also no information referrals requested in FC, JP, or JC. Discussion Internet users join online health support communities (such as those available on MedHelp.org) even though a plethora of alcohol abuse and other health information is available on the Internet with via vetted information sources. These e-patients were looking for the social component (peer support) most likely because they were not health professionals and wanted to be in a place where they could use laymen terms to share insights with each other as opposed to purely health information sites (e.g., WebMD), where users may experience difficulty in understanding the large quantity of information available. The added social components in support communities—where e-patients can have their questions answered and hear other e-patients’ experiences—provide more easily digestible information, for example, advice about applying new lifestyle changes. Although people can obtain social support from their existing offline social networks, participation in online support groups has added benefits. We identified different types of informational support in the MedHelp alcoholism community across three text-based CMC formats. Each format was used differently. Forums were used for asking for and sharing information with a wider audience. Journal comments were similar to notes with smaller groups of individuals interacting, which might be an indication that journal readers consider the author as a friend. Notes were not the main source for seeking information but, rather, for maintaining relationships. Users joined the community seeking information but potentially remain active because of relationships formed with other members in the community. Our study is different from related studies of the same type concerning the support identified among usergenerated content: (a) We observed and provided results for both offered and requested support, and (b) we investigated three CMC formats (Bambina, 2007; Braithwaite et al., 1999; Coursaris & Liu, 2009; Cunningham et al., 2008; Eichhorn, 2008; McCormack, 2010; Pfiel & Zaphiris, 2007; Preece, 1999). Some previous studies reported an overall amount of informational support given without presenting the specific subtypes (Bambina, 2007; Coursaris & Liu, 2009; Eichhorn, 2008). We present our results for overall informational support offered along with the five subtypes. Most of these previous studies only investigate the social support on one CMC format (i.e., forums). Our study takes a further step to study the social support on journals and notes in social media. In terms of the results, we found that fact and advice were the most frequently exchanged subtypes, which is not quite the same as some previous studies (Lau & Kwok, 2009). Some previous studies found that personal experience was exchanged the most in the online forums (Cunningham et al., 2008; Lau & Kwok, 2009; Pfiel & Zaphiris, 2007; Preece, 1999). We found that fact was always highest in both

offered and requested dimensions, which agrees with one other study (McCormack, 2010). There are several plausible explanations for the variance between our study and other studies. First, this study collected data from different text-based communication formats (journal, notes) than previous studies (mailing lists, discussion boards). The architectural elements are different and can affect communication. Second, the members of the MedHelp community are allowed to and often communicate with each other across multiple CMC formats instead of just one format (i.e., e-mail lists). Features such as the profile page and journals are similar to providing rooms for people to talk about more specific things and have fewer interruptions, and this availability impacts the conversations on the communal areas to be more formal and the other areas to be less so. Third, the space formed from these three different CMC formats contains conventions or social norms that influence any new member’s contributions to the community with that particular format. Although early adopters of this website may have been influenced by the features provided in the SNS (Papacharissi, 2009), new members would likely stay quiet until they could grasp the feel of the community before posting messages. To the e-patient, the interactivity of an online community is different from perusing static information pages because of the added social component. A forum space is similar to a waiting room at the clinic in that people know it is more public than the doctor’s office. In terms of informational and emotional content exchanged in the community, users were selective in what they write and with whom they interacted across the CMC formats. In the forums, it appeared that they treated the space as a Q&A forum, whereas on profile pages and journals the personal nature may explain their behavior in exchanging content that is more emotional. It may be the environment that makes them behave this way, yet it may also be that they want to behave this way and perceive those environments to be proper for that type of conduct. Our findings on the informational support exchange patterns identified across the three CMC formats in the MedHelp alcoholism community showed that offered support had three main patterns. Journal comments were similar to notes, in that most messages offered advice, with opinions a close second. The next were personal experiences or information referrals. We consider these to be a less formal means of communication because users know each other and because they tend to be one on one interactive format. Another pattern we found among offered support was journal comments and forum comments similarly having mostly advice, and then an approximately equivalent number of messages offering opinion, personal experience, or referrals. Surprisingly, journal posts were not similar to forum comments. JP offered more referral (R > O/P > A) than FC, which had more advice (A > O/P/R). In the results showing requested support types, the samples of forum posts, forum comments, and journal posts were similar in the aspect of showing exchanges of mostly facts with next ranking support type of advice, and thirdly opinions. Notes

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

49

and journal comments almost exclusively asked for facts. Perhaps users were checking up on friends and wanted to know how they were doing. We hypothesized that these pattern differences could be a result of the CMC format, but these differences can also be due to the purpose of communication. CMC format structural differences. The three formats in this study (forum, journal, and notes) were different from the settings of the other studies (e-mail lists and discussion boards). Even among these three, there are some differences in accessibility. Forums are more easily accessible because of the linkage from the main page, the places where update notifications are listed, and the access to the space. Journals are different from forums. Journals are similar to writing thoughts and feelings in “diaries”. In a way, when the writer presents his or her content, it does not necessarily have to be for requesting information. Instead, by being the author of a journal post, he or she can initiate a discussion without the use of questions. He or she can be the owner of the discussion. Journals are moderated more than forums because a journal belongs to a user, and the user can control who reads the posts and can delete comments. The notes format is one-to-one communication so people tend to use it for more social reasons. These three formats (forum, journal, and notes) are quite common on social media platforms including profile pages, especially for these targeted interest groups. They differ from previous studies that use e-mail lists in that e-mail lists do not contain profile information because they are not linked to a profile page. It is a different way of representing a user’s identity and having a “front door” to that identity. SNS tend to focus more on relationships (especially the “friends” listed on a profile page) and interpersonal communication with the plethora of messaging features. Last, it is different because members can post public notes to each other that branch off from the main community. Communicating with intent. The MedHelp communities have several communication formats, each used for a different purpose. Constructing an arena for conversation (i.e., e-mail lists and bulletin boards) is good for group style; however, for more tailored and personal communication between smaller groups (i.e., more focused topic) or between two individuals, the other formats (journal and notes) are more suitable. People have different reasons for participating in an online community; for example, some members sought information, whereas others sought compassion and intimacy (Chuang & Yang, 2012; Rau et al., 2009). In addition, patients may go through waves of information needs (Radin, 2000). Posting to the forum may be for a different purpose than posting to a journal or notes; for example, one may disclose personal information as a strategy for finding tailored information or to document experiences. Social media make it easy to share and seek information from peers. Internet users may also want to stay in touch 50

with close friends and family (Gilbert & Karahoulios, 2009; Kovic et al., 2008). Often in the case of visiting online health communities, however, visitors are searching for people who have experienced similar situations and hear stories, and practical advice tailored to their specific situation (Overberg et al., 2010; Wright & Bell, 2003). This is where SNS’s features for showing a message to a small circle of recipients can impact the content and wording within a message. Networked private space. Over time, new members of a community may adopt the same behaviors as the early adopters, especially if there are no obvious signals (Donath, 2007; Meyrowitz, 1987). Studies support this theory, as SNS users tend to communicate one-on-one or within small groups (Takahashi et al., 2009). Although the CMC format itself can influence different behavior patterns based on the user’s perception of the space, sometimes users also have different intents while selecting a format. Notes and forum comments are similar to journal comments, but forum posts stand out as having a different pattern than other formats. Because privacy can be controlled through notes and journals, they are more personal than the public forum. It is possible that the users did not find it necessary to use privacy controls for journals and notes because it appears more private. In addition, in a physical setting, it is easy for one to perceive the relative privacy of the space. However, in an online environment, the amount of privacy is not as transparent. In this case, perhaps the MedHelp users do not assess the online setting as they would a physical face-to-face setting. In light of the content observed through this community (i.e., blackouts, possible violent episodes, etc.), the online setting diminishes the amount of stigma that would be present in face-to-face support. Reported results of the levels of informational support types along presented theories point to location of communication elements as something designers should pay attention to when designing an online community. Group style communication can be encouraged with a design that has the style of a bulletin board where users can easily view previous threads of communication and have increased accessibility to all the content. Interpersonal communication between two users or smaller groups with a more focused topic, which can be between two individuals, can be encouraged through formats that show one visible “ownership” of space such as journals and notes. These are also only accessible by crossing through someone’s profile page. Future Work There were a few limitations in this work. Future work can identify patterns at the message level rather than patterns for each sample so as to better understand the construction of each message. This study did not account for members’ support profiles such as whether they are more likely to

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

provide or request support. Future work that identifies whether support was given explicitly to a recipient could be useful in understanding the helpfulness of participating in an online support community. Future research could also improve on the traditional use of content analysis and incorporate a more inductive approach to the analysis. Traditional use of content analysis could include analyzing one coding scheme per variable rather than one overall scheme so as to provide specific coding reliability for each measure and more rigor in the results. The inductive approach to content analysis would open the door to additional insight into members of the MedHelp alcoholism community to better understand their purposes and the uses of each CMC format. More specifically, this approach could shift from a focus on messages on a technology platform to a patient end-user focus, giving a more comprehensive view of information and communication behavior. This would be directly related to guidelines that can be constructed for designers because the data would show how users behave outside of the intended design. Conclusion This study, as part of a larger, multipart study, investigated the informational support exchanged across different software features of a health-related social networking site. We identified different types of information exchanged as social support, with a different pattern on each CMC format. Although e-patients can seek information from validated health information websites or health content sites such as WebMD, they also participate in support groups for help with coping with chronic health conditions. We identified different types of informational support in the MedHelp alcoholism community across three textbased CMC formats. Each format was used differently. Forums were used for asking and sharing information with a wider audience. Journal comments were similar to notes with smaller groups of interacting individuals, which may be an indication that journal readers consider the author a friend. Notes were not the main format for seeking information from the community at large but, rather, for friends to communicate with each other using brief messages. Our results for the different levels of support types on each format suggest the need for further analysis of the timeline of user behavior from when a user first posts a message to later messages as the user becomes more familiar with the community. For example, users may have initially joined the community seeking information, however, at later stages of information gathering they use the notes for targeted questions. References Ahn, Y., Han, S., Kwak, H., Moon, S., & Jeong, H. (2007). Analysis of topological characteristics of huge online social networking services. In Proceedings of the 16th International Conference on World Wide Web (WWW ’07; pp. 835–844). New York: ACM. DOI: 10.1145/ 1242572.1242685

Alcoholics Anonymous. (2001). Alcoholics anonymous, 4th edition. New York: A.A. World Services. Ancker, J.S., Carpenter, K.M., Greene, P., Hoffman, R., Kukafka, R., Marlow, L.A.V., . . . Quillin, J.M. (2009). Peer-to-peer communication, cancer prevention, and the Internet. Journal of Health Communication: International Perspectives, 14(1 supp 1), 38–46. Bambina, A.D. (2007). Online social support: The interplay of social networks and computer-mediated communication. Cambria Press. Beattie, M.C., & Longabaugh, R. (1999). General and alcohol-specific social support following treatment. Addictive Behaviors, 24(5), 593– 606. Boase, J., Rainie, L., Horrigan, J., & Wellman, B. (2006). The strength of Internet ties. Washington, DC: Pew Internet & American Life Project. p. 52. boyd, D.M., & Ellison, N.B. 2007. Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13(1), article 11. Braithwaite, D.O., Waldron, V.R., & Finn, J. (1999). Communication of social support in computer-mediated groups for people with disabilities. Health Communication, 11(2), 123–151. Chuang, K., & Yang, C.C. (2012). Interaction patterns of nurturant support exchanged in online health social networking. Journal of Medical Internet Research, 14(3), e54. doi:10.2196/jmir.1824 Civan, A., & Pratt, W. (2007). Threading together patient expertise. In AMIA 2007 Proceedings Symposium (pp. 140–144) Bethesda, MD: American Medical Informatics Association. Cohen, S. (2004). Social relationships and health. American Psychologist, 59(8), 676–684. Coursaris, C.K., & Liu, M. (2009). An analysis of social support exchanges in online HIV/AIDS self-help groups. Computers in Human Behavior, 25(4), 911–918. Cunningham, A.J., Humphreys, K., Kypri, K., & van Mierlo, T. (2006). Formative evaluation and three-month follow-up of an online personalized assessment feedback intervention for problem drinkers. Journal of Medical Internet Research, 8, e5. Retrieved from http://www.jmir.org/ 2006/2002/e2005/ Cunningham, J.A., van Mierlo, T., & Fournier, R. (2008). An online support group for problem drinkers: AlcoholHelpCenter.net. Patient Education and Counseling, 70(2), 193–198. Cutrona, C.E., & Suhr, J.A. (1992). Controllability of stressful events and satisfaction with spouse support behaviors. Communication Research, 19(2), 154–174. Donath, J. (2007). Signals in social supernets. Journal of ComputerMediated Communication, 13(1), article 12. Retrieved from http:// jcmc.indiana.edu/vol13/issue1/donath.html Eichhorn, K.C. (2008). Soliciting and providing social support over the Internet: An investigation of online eating disorder support groups. Journal of Computer-Mediated Communication, 14(1), 67–78. Ellison, N., Lampe, C., & Steinfield, C. (2009). Social network sites and society: current trends and future possibilities. Interactions Magazine, 16(1). Fox, S., & Jones, S. (2009) The social life of health information. Americans’ pursuit of health takes place within a widening network of both online and offline sources. Retrieved from http://www.pewinternet.org/ Reports/2009/8-The-Social-Life-of-Health-Information.aspx Gilbert, E., & Karahalios, K. (2009, April). Predicting tie strength with social media. Paper presented at the Proceedings of the 27th International Conference on Human factors in Computing Systems. Boston, MA. Retrieved from http://dx.doi.org/10.1145/1518701.1518736 Høybye, M.T., Johansen, C., & Tjørnhøj-Thomsen, T. (2005). Online interaction: Effects of storytelling in an Internet breast cancer support group. Psycho-Oncology , 14(3), 211–220. Humphreys, L. (2007) Mobile social networks and social practice: A case study of dodgeball. Journal of Computer-Mediated Communication 13(1). Retrieved from http://jcmc.indiana.edu/vol13/ issue1/humphreys.html Joinson, A.N., & Banyard, P. (2003). Seeking alcohol information on the Internet. ASLIB Proceedings, 55, 313.

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi

51

Klaw, E., Huebsch, P.D., & Humphreys, K. (2000). Communication patterns in an on-line mutual help group for problem drinkers. Journal of Community Psychology 28(5), 535–546. Linke, S., Murray, E., Butler, C., & Wallace, P. (2007). Internet-based interactive health intervention for the promotion of sensible drinking: Patterns of use and potential impact on members of the general public. Journal of Medical Internet Research, 9, e10. Retrieved from http://www .jmir.org/2007/2002/e2010/ Lofland, J., & Lofland, J. (2006). Analyzing social settings: A guide to qualitative observation and analysis. Belmont, CA: Wadsworth/Thomson Learning. Lau, A.S., & Kwok, T.Y. (2009). Social features in online communities for healthcare consumers—A review. In A.A. Ozok & P. Zaphiris (Eds.), Online communities and social computing (Vol. 5621, pp. 682–689). Heidelberg, Germany: Springer. McCormack, A. (2010). Individuals with eating disorders and the use of online support groups as a form of social support. Computers, Informatics, Nursing, 28(1), 12–19. McKenna, K.Y.A., Green, A.S., & Gleason, M.E.J. (2002). Relationship formation on the Internet: What’s the big attraction? Journal of Social Issues, 58(1), 9–31. McLure Wasko, M., & Faraj, S. (2000). “It is what one does”: Why people participate and help others in electronic communities of practice. Journal of Strategic Information Systems, 9(2–3), 155–173. Meyrowitz, J. (1986) No sense of place. New York: Oxford University Press. Oldenburg, R. (1989). The great good place: Cafes, coffee shops, community centers, beauty parlors, general stores, bars, hangouts, and how they get you through the day. New York: Paragon House. Overberg, R., Otten, W., de Man, A., Toussaint, P., Westenbrink, J., & Zwetsloot-Schonk, B. (2010). How breast cancer patients want to search for and retrieve information from stories of other patients on the Internet: an online randomized controlled experiment. Journal of Medical Internet Research, 12, e7. Retrieved from http://www.jmir.org/2010/2011/e2017/ Papacharissi, Z. (2009). The virtual geographies of social networks: A comparative analysis of Facebook, LinkedIn and ASmallWorld. New Media & Society, 11(1–2), 199–220. Pfeil, U., & Zaphiris, P. (2007). Patterns of empathy in online communication. Paper presented at the Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. Preece, J. (1999). Empathy online. Virtual Reality 4(1), 74–84.

52

Preece, J. (2000). Online Communities: Designing Usability, Supporting Sociability. New York: John Wiley. Putnam, R.D. (1995). Bowling alone: America’s declining social capital. Journal of Democracy 6(1), 65–78: doi: 10.1353/jod.1995.0002 Radin, P.M. (2001). “Hello, my friends”: Social capital and transformations of trust in a grassroots breast cancer Website. Washington, University of Washington: 141. Rau, P.-L.P., Gao, Q., & Ding, Y. (2008). Relationship between the level of intimacy and lurking in online social network services. Computers in Human Behavior, 24(6), 2757–2770. Rosenquist, J.N., Murabito, J., Fowler, J.H., & Christakis, N.A. (2010). The spread of alcohol consumption behavior in a large social network. Annals of Internal Medicine, 152(7), 426–433. doi: 10.7326/0003-4819-152-7201004060-00007 Smith, K.M. (2004). “Electronic eavesdropping”: The ethical issues involved in conducting a virtual ethnography. Online Social Research; Methods, Issues, & Ethics, 223–238. Takahashi, Y., Uchida, C., Miyaki, K., Sakai, M., Shimbo, T., & Nakayama, T. (2009). Potential benefits and harms of a peer support social network service on the internet for people with depressive tendencies: Qualitative content analysis and social network analysis. Journal of Medical Internet Research, 11, e29. Retrieved from http://www.jmir.org/2009/2003/e2029 Thelwall, M., & Wilkinson, D. (2010). Public dialogs in social network sites: What is their purpose? Journal of the American Society for Information Science and Technology, 61(2), 392–404. Tufekci, Z. (2008). Can you see me now? Audience and disclosure management in online social network sites. Bulletin of Science and Technology Studies. 11(4), 544–564. Walstrom, M. (2004). Seeing and sensing online interaction: An interpretive interactionist approach to USENET support group research. Online Social Research: Methods, Issues, & Ethics, 81–97. Winzelberg, A. (1997). The analysis of an electronic support group for individuals with eating disorders. Computers in Human Behaviour, 13, 393–407. World Health Organization (2010, May, 21). Call for action to reduce harmful use of alcohol. Retrieved from http://www.who.int/mediacentre/ news/releases/2010/alcohol_20100521/en/ Wright, K.B., & Bell, S.B. (2003). Health-related support groups on the Internet: Linking empirical findings to social support and computermediated communication theory. Journal of Health Psychology, 8(1), 39–54. doi: 10.1177/1359105303008001429

JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY—January 2014 DOI: 10.1002/asi