Content-Type: text/html

                On-line Support for Eating Disorders


Running head: ON-LINE SUPPORT FOR EATING DISORDERS



Soliciting and Expressing Social Support Over the Internet:
 An Investigation of On-line Eating Disorder Support Groups





Kristen L. Campbell
Doctoral Student
School of Communication
University of Miami
Frances Wolfson Building
5100 Brunson Drive
Coral Gables, FL, 33146
Phone: 305-284-2034
 Fax: 305-284-5216
 E-mail: <a href="/cgi-bin/wa?LOGON=A3%3Dind0209B%26L%3DAEJMC%26E%3D8bit%26P%3D2707712%26B%3D--%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D_24987201%253D%253D_%26T%3Dtext%252Fplain%3B%2520charset%3Diso-8859-1" >[log in to unmask]</a>





Paper submitted for consideration to the Association for Education in Journalism and Mass Communication, Mass Communication and Society Division, Miami Beach, FL, April 1, 2002



Abstract
Using a longitudinal and systematic sample of 490 postings, this study analyzed the themes, the type of social support, and the strategies used to solicit social support provided on the top five Yahoo! eating disorder discussion boards.  Optimal matching theory, the notion that it is possible to attribute stressful events to social supportive behavior in order to find the most favorable match, led to the formulation of two hypothesis. Both were supported.  First, messages seeking informational support were expected to be more prevalent than those seeking instrumental support. Second, messages posted on these on-line channels were expected to be more emotionally supportive than informationally supportive. The findings also reveal that sharing experiences is the most frequent strategy for soliciting support. Moreover, this study revealed that the most frequent theme addressed on the discussion board is positive affect, or providing encouragement and constructive feedback. Finally, f
indings indicate support for the optimal match theory by suggesting that those individuals suffering from an eating disorder are seeking informational support.
 Soliciting and Expressing Social Support Over the Internet:
An Investigation of On-line Eating Disorder Support Groups
Anorexia Nervosa and Bulimia Nervosa are the two most common eating disorders among North American women between the ages of 16 and 25. Anorexia Nervosa occurs in about 5-10 out of 1000 women while Bulimia Nervosa occurs in about 20-50 out of 1000 women (Crowther, Wolf, & Shepard, 1992; Hsu, 1996). These two eating disorders are associated with a distorted perception of body shape and weight. According to the American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R), the diagnostic criteria for Anorexia Nervosa are: a) refusal to maintain body weight over a minimal normal weight for age and height; b) intense fear of gaining weight or becoming fat, even though underweight; c) absence of at least three consecutive menstrual cycles in females. The association defines the diagnostic criteria for Bulimia Nervosa as: a) recurrent episodes of binge eating; b) feeling of lack of control over eating behavior during binges; c) recourse to self-indu
ced vomiting, use of laxatives or diuretics, strict dieting or fasting, or vigorous exercise in order to prevent weight gain; d) minimum average of two binge eating episodes a week for at least three months; and e) persistent concern over body shape and weight.
Eating disorders carry a dramatic social stigma. Those individuals suffering from an eating disorder only choose to disclose their condition to select people, which limits their overall social support networks. More recently, those suffering from eating disorders have turned to Internet sites for anonymous social support.  Internet sites provide information, guidance, and support groups to meet other individuals suffering from similar symptoms. On-line support groups usually take the form of listservs, where members are able to write and/or receive messages through email, or discussion boards, where members are able to read and/or post messages to the specific site. The present study will focus on on-line support groups, through the use of discussion boards. However, little is known about the participants or their approach to this new medium. With the absence of nonverbal cues on the Internet, it is difficult to utilize traditional forms of social support behaviors, such as listening, sympathy, and concern for on-line support group members.
Before focusing on cultivating successful supportive relationships on-line and developing strategies on how to provide support through computer-mediated communication, it is critical to gain a greater understanding of the existing social supportive behaviors.   By analyzing the current supportive behaviors existing in on-line eating disorder discussion boards, researchers could offer greater insights into the types of social support that eating disordered individuals are seeking and the types of support that others are providing.
This study will analyze the themes, the type of social support, and the strategies used to solicit social support posted on the top five Yahoo! on-line eating disorder discussion boards. In addition, it will link theory and content by relying on the optimal match theory to predict the types of messages that individuals with eating disorders are seeking. Furthermore, it will analyze message postings longitudinally and systematically in order to gain a more valid perspective. This study will begin by explaining the theoretical framework and reviewing the relevant literature.


Theoretical Framework
Optimal match theory suggests that it is possible to attribute stressful events to social supportive behavior in order to find the most favorable match (Cutrona & Russell, 1990). The theory argues that social support is multidimensional and can produce optimal behavior when an individual is confronted with a stressful event. According to the optimal match theory all stressful events, including eating disorders, can be integrated and simplified into four dimensions: desirability, controllability, duration of consequences, and life domain. Using these markers of psychological consequences, Cutrona and Russell (1990) built a theory of "optimal stress-social support matching based on the understanding of the psychological processes stimulated by different classes of events" (p. 327).
The four dimensions of stress are described as follows. Desirability refers to the extent to which a stressful event involves either a potential or actual gain or a potential or actual loss.  In other words, a desirable stressful event would involve a gain or growth and an undesirable event would involve a loss. As Cutrona and Russell (1990) point out,  "The major difference between desirability and undesirable stressful events seems to be the nature and intensity of the negative emotions they engender. Desirable events in which stress derives from uncertainty over whether or not a desired goal can be achieved are associated with feelings of anxiety, whereas undesirable events have shown to produce depression" (p. 329). An eating disorder would be considered an undesirable event because it involves a distorted perception of self, depression, and a sense of loss of control. According to the optimal match theory, more support will be needed to deal with an undesirable stressor than
with a desirable stressor.
Next, whether an event is considered controllable or uncontrollable will determine what type of social support is needed. Controllability refers to the extent to which an individual has control over the outcome. This is the most influential dimension. Cutrona and Russell (1990) contend that "uncontrollable events (harm or loss) will require social support components that foster emotion-focused coping and controllable events (threat or challenge) will require social support components that foster problem-focused coping" (p. 329).  Emotion-focused coping involves the opportunity to ventilate emotions, reevaluate the severity of one's loss, or experience positive emotions that derive from sources not lost because of the stress (e.g., reminders that one is loved). On the other hand, problem-focused coping deals with advice, information, feedback on an individual's plans for action, actual assistance, or support that reinforces an individual's competence level. Furthermore, they predict that instrumental (tangible) support will be more effective following uncontrollable events and informational (advice or guidance) support will be more effective following controllable events. This makes sense because instrumental support will be used to help replace the resource or item that was lost and informational support will be used to help prevent or solve problems. Eating disorders would be considered a controllable event because the individuals have control over the onset and termination of the stressful event.
Another stress dimension that is related to control is the duration of the consequence. Cutrona and Russell (1990) suggest that "The longer the effects last, the more emotional support will be required to bolster morale and the more tangible support will be required, because need persists over a longer period of time" (p. 330). It is difficult to determine the impact of eating disorders because it involves rooted psychological factors, such as a distorted perception of self, depression, and low self-esteem. These factors are embedded in the psyche and are not restored as soon as the behavior is corrected.
Finally, life domain is a dimension of a stressful event that considers the type of loss that is experienced, in order to establish a replacement. Optimal match theory recognizes four types of losses: (1) assets, (2) relationships, (3) achievement, and (4) social roles.  An eating disorder would be considered a loss of achievement because it involves a disappointed self, due to lack of control. Cutrona and Russell (1990) claim that "When the loss of threat is primarily in the achievement domain, reassurance of capabilities and worth would most directly address the individual's deficit" (p. 330). Overall, this theoretical model suggests that desirability, controllability, duration of consequence and life domain are the dimensions of life events that determine strains and challenges, which then determine which types of social supportive behaviors will most likely lead to healthy outcomes. After reviewing the literature, two hypotheses will be formulated based on the optimal match th
eoretical model.
Review of Literature
        In order to examine the types of messages solicited and provided within computer-mediated discussion boards, it is necessary to review the social support and support group literature.
Social Support
One way to control the onset and development of Anorexia Nervosa and Bulimia Nervosa is through multifaceted prevention. Primary prevention refers to attempts to reduce the occurrence of new cases of eating disorders. Secondary prevention refers to the early identification of individuals at risk of an eating disorder. Finally, tertiary prevention involves the prompt identification and treatment of individuals who have a full-blown eating disorder (Kessler & Albee, 1975; Piran, Levine, & Steiner-Adair, 1999).  Social support can prove to be influential in all three domains of prevention.  An effective social support network can provide anything from a positive body image to healthy eating habits and appropriate resource material.
Social support has received a great deal of attention in the psychological literature. Researchers have argued that it is multidimensional and they have presented a variety of theoretical models to help understand this construct (Cobb, 1979; Cohen, Mermelstein, Kamarck, & Hoberman, 1985; Kahn, 1979; North, 1997; Schaefer, Coyne, & Lazarus, 1981; Weiss, 1974). More recently, communication scholars have contributed to the area of social support (Albrecht & Adelman, 1984, 1987; Albrecht, Burleson, & Goldsmith, 1994; Albrecht, Burleson, & Sarason, 1992; Burleson, 1994; Rook, 1995; Wright, in press; Zelley, 2001; Zimmerman & Applegate, 1994).  Burleson et al. (1994) argue that social support should be studied through the communication discipline because it ultimately is conveyed through the interaction between individuals.  They suggest that social support is created and sustained through the context of relational communication.
Researchers have agreed that emotional support is an important component of social support. However, emotional support is defined in a variety of ways. Buhrmester, Furman, Wittenberg, and Reis (1988) defined emotional support as providing comfort to others when they are experiencing problems or distress. Weber and Patterson (1996) argue that emotional support consists of strategies that elicit caring such as listening, nonverbal immediacy, displays of caring, and the offering of advice. Cohen and Wills (1985) label emotional support as expressions of concern, compassion, sympathy, and esteem for another individual. Hays, Magee, and Chauncey (1994) characterize emotional support as feelings of comfort and reassurance that the recipient is loved, understood, and respected.  Finally, Hill (1987) asserts that expressions of sympathy and nurturing are examples of emotionally supportive messages. Therefore, it is easy to imagine demonstrating emotional support in face-to-face interactio
ns, but it is difficult to envision how this can be done on-line without the presence of nonverbal cues.
Those individuals who are suffering from eating disorders have fewer people to turn to for support than individuals who are not.  Zelley (2001) compared the social support networks for women with eating disorders (ED) and women without eating disorders (nonED). Their results indicate that those women who suffered from disordered eating perceive fewer friends in their social support network than those women who did not suffer. Another interesting finding addressed the perception of these social support networks. They found no differences between ED women and nonED women in the supportiveness, helpfulness, and sensitivity of these social support systems. In other words, women suffering from an eating disorder identified fewer people providing support but these people were not perceived as less effective, compared to those identified by nonED women.  Zelley (2001) suggests:
_more attention should focus on the relationship between ED women and their friendship network as well as ways in which to cultivate these relationships rather than ignore, avoid, or destroy them through interpersonal incompetence and or ignorance. Perhaps, it is not information on diet, body image, and self-esteem that helps prevent ED onset and facilitate successful recovery but rather, knowledge of communication competence in terms of soliciting and receiving adequate social support that provides this inoculation. (p. 12)
In the same vein, Rorty, Yager, Buckwalter, and Rossotto (1999) compared individuals actively bulimic (ABN), recovering form bulimia (RBN), and those individuals who have no history of an eating disorder. They found that the ABN group identified significantly fewer individuals who were able to provide emotional support than did the other groups. However, there were no significant differences in terms of the number of individuals identified to provide informational support or advice between groups. In addition, both active and recovering bulimic groups were dissatisfied by the quality of emotional support received by relatives.
Support Groups
Studies indicate that there are many social, emotional, and physical benefits to those individuals who participate in support groups. Magen and Glajchen (1999) found that members of cancer support groups identified cohesion as the most important therapeutic factor. Cohesion is crucial in support groups because it allows members to develop emotional bonds and permits self-disclosure (Yalom, 1995). However, there are many differences between on-line and face-to-face support groups.
Comparison of on-line and  face-to-face support groups.  Sparks (1992) contends that "Unlike face-to-face support groups, electronic support groups offer 24-hour availability, selective participation in entering and responding to messages, anonymity and privacy, immediate and/or delayed responding, and recording of transactions" (p. 62). He also notes that on-line support groups allow individuals to gather worldwide without incurring the cost of traveling. Several researchers have addressed the advantages and disadvantages of on-line support groups (Lemm, Reppert, & Visich, 1998; Walther, 1996). Researchers have also suggested that as people spend more time on-line and become comfortable, they will compensate for the absence of nonverbal cues by using creative text-based communication (Matheson & Zanna, 1988; Walther, 1996; Walther & Burgoon, 1992). Over time, on-line relationships are expected to resemble face-to-face relationships as individuals compensate for the medium and use it more frequently (Walther & Burgoon, 1992).
 Lin and Peek (1999) argued that the provision of emotional support is affected by the size and density of an individual's support network. People with larger and denser networks have been found to be more likely to receive support, and emotional support has been empirically linked to mental health outcomes such as lower perceived stress and the ability to cope with problems better. For people seeking support through on-line groups, there is an opportunity to develop larger support networks than would be possible through face-to-face interaction. In addition, even networks of people who are relatively strangers (which is often the case initially in on-line groups) have been found to provide emotional support (Adelman, Parks, & Albrecht, 1987).
        On-line support groups.  Sharf (1997) examined a breast cancer on-line discussion group and identified advantages and disadvantages to providing on-line support. She admits that "While the inability to offer a hug or understanding eye contact is a limitation in interpersonal responsiveness, those same conditions may also encourage a more uninhibited outpouring of feelings without conscious of tears, a disfigurement, or other physical barrier" (p. 74). Also, Sharf suggests that in order to overcompensate for the loss of nonverbal cues, on-line participants are organizing meetings to unite members face-to-face.
Wright (2000) explored the relationship between social support satisfaction and perceptions of stress among participants in an on-line support community for older adults. The results indicated a modest relationship between high support satisfaction and perceptions of lower stress.
Furthermore, Lemm, Reppert, and Visich (1998) conducted a content analysis to determine how on-line cancer support group members use the Internet. They identified eight major themes for posting messages: 1) Information giving/seeking; 2) Personal opinions; 3) Encouragement/Support; 4) Personal experiences (patients/caregivers); 5) Thanks; 6) Humor; 7) Prayer; and 8) Miscellaneous.
Finally, Turner, Grube, and Meyers (2001) examined the extent to which non-computer mediated support systems influence the potential of individuals to develop hyperpersonal relationships within an on-line cancer support group. They found a nonsignificant relationship between perceptions of the on-line cancer Listserv and perceptions of their face-to-face partner regarding support. Therefore, the authors suggest that the computer-mediated support is an important function of support. In addition, they found that participation was high on the on-line support group only when they perceived that the depth and support that they were receiving from their face-to-face interaction was low. Using optimal matching theory as a theoretical framework, Turner et al. (2001) suggest that "the listserv may have provided these participants with a 'match' of some aspect of support and understanding concerning their illness that was missing from the face-to-face partner" (p. 246).
Hypotheses and Research Questions
This study also utilized optimal match theory to examine the content of on-line eating disorder support groups. The following hypotheses were tested:
H1: There will be more messages seeking informational support on the top five Yahoo! eating disordered discussion boards than those seeking instrumental support.
H2: Informational support will be the most frequent type of social support provided on the top five Yahoo! eating disordered discussion boards.
Furthermore, this study examined how individuals solicit support and the frequency of these approaches. North (1997) identified five strategies that individuals with eating disorders use to solicit social support on-line. These strategies include:
(1) self deprecating comments, (2) shared experiences, (3) requests for information, (4) statements of personal success, and (5) statements of extreme behavior. The following research question was posed:
RQ1:    What is the most frequent strategy that individuals with eating disorders use on-line to solicit social support?
Finally, this study addressed the frequency of  themes that individuals with eating disorders discuss. Winzelberg (1997) identified five themes generated within eating disorder support groups. They include:  (1) coping with external pressures from family and friends, (2) reactions to the cultural pressures to be thin and standard of beauty portrayed in the media, (3) reminiscence of the psychological symptoms of bingeing, purging, and starvation, (4) negative affect, and (5) recommendations for psychological treatment and reflections on the benefits members had received from such treatment. The following research question was posed:
RQ2:    What is the most frequent theme addressed within on-line eating disorder
discussion boards?
Method
Sampling
        Because there are virtually hundreds of on-line eating disorder support groups on the Internet, the universe of this content analysis was limited to the top five most popular Yahoo! eating disorder discussion boards in 20011.  Other than manageability, there were several reasons for delineating the population in those terms. First, both Nielsen/Net Ratings and Jupiter Media Metrix, the leading user-based ratings services on the web, identified Yahoo! as having the largest audience reach than any other searching tool (Sullivan, 2001). Furthermore, StatMarket, a search engine ratings which break out search-specific traffic from other traffic, such as retrieving email or checking a horoscope, rated Yahoo! as the most popular search engine (Sullivan, 2001). Hence, it can be argued that Yahoo! is a popular site that attracts a variety of users.  In addition, user profiles were available on the Yahoo! site and could be used to gain insight into the demographic characteristics of those
individuals who post messages on-line. Finally, all messages on the Yahoo! site were posted, archived, and easily accessible for a systematic sampling approach. With this in mind, the Yahoo! site was chosen for this study.
The number of members per support group determined the top five most popular discussion boards: (1) Bulimiasupportgroup, (2) BOA,  (3) Maleseatingdisorders (4) Dieana (5) Bulimiaandanorexianetherworld. Data were gathered during the week of March 17-23, 2002. Only those groups that were in working condition during the data collection week were considered. Message postings were listed in chronological order and were selected using a skip interval: 101 postings for Bulimiasupportgroup; 100 postings for BOA; 97 postings for Maleseatingdisorder; 99 postings for Dieana; and 93 postings for Bulimiaandanorexianetherworld.
Coding Procedures
        Unit of analysis.  The unit of analysis for this study was the individual message posting. However, in order to categorize the soliciting strategies and social supportive messages the sample frame consisted of grammatical proposition of phrases. The variables examined include themes, strategies, and type of social support.
Operational Definitions.  A single theme of each of the discussion board postings was coded into eight categories.  Six of these categories originated from Winzelberg (1997): 1) coping with weight,  2) coping with external pressures from family and friends,  3) reactions to the cultural pressures to be thin and standard of beauty portrayed in the media, 4) recalling symptoms of binging, purging, and starvation, 5) negative affect, 6) recommendations for prosocial behavior (i.e., treatment). Three additional themes were added to account for more of the messages: 1) newcomer message, 2) positive affect, and  3). unrelated to eating disorders.  Strategies that were used to solicit social support were coded into the following five categories developed by North (1997): 1) self-deprecating comments, 2) shared experiences, 3) requests for information, 4) statements of personal success, and 5) statements of extreme behavior. The type of social support was coded into the following five categories:  1) informational, 2) tangible (instrumental), 3) esteem, 4) emotional and 5) network support.  Messages were also coded for gender, marital status (single, long-term relationship, married, not determined), location (United States, outside the United States, not determined), time (6:01am-12:00pm, 12:01-6:00pm, 6:01pm-12:00am, 12:01am-6:00am, not determined), and age.
Intercoder Reliability
        A subsample of the coding (15%) was randomly selected and recoded by a trained, independent graduate student. Computation of nominal data was based on Cohen's Kappa (1960), which accounts for agreement by chance, was used to compute the following intercoder reliability figures: group, .98; primary theme, .86; primary soliciting strategy, .85; primary form of support, .83; gender, .91; marital status, .96; location, .96; and time, .93. Holsti's (1969) formula was used to compute intercoder reliability for age, .92.
Results
User profile
Approximately 37% of the individuals posting messages on the top five Yahoo! support groups were between the ages of 20 and 30. Sixteen percent of the individuals were between the ages of 13 and 19, and 11% were older than 30. Females dominated the on-line support group postings, authoring 82% of the sample messages. Males posted about 14% of the messages. Of the known marital status of the users, 40% reported to be single, 11% reported to be in a long-term relationship and 9% were married. Of the known location of the users, 40% of the message postings originated from individuals who lived inside the United States, while about 17% came out outside the United States.  Moreover, 61% of the messages were posted between 12:01pm and Midnight.

Hypotheses and Research Questions
Hypothesis 1 predicted that there would be more messages seeking informational support than instrumental or tangible support. Of the messages, about 17% of the messages provided informational support and about 7% of the messages provided instrumental support (see Table 1). Thus, hypothesis 1 was supported. A chi-square goodness of fit (_2 (5, N= 490) = 299.67, p < .001) which tested for differences between observed and expected frequencies of the social support strategies was statistically significant and indicated support for the opposite prediction.
Hypothesis 2 predicted that informational support would be the most frequent type of social support provided. Informational support was the most frequent type of support provided with about 17%. Thus, hypothesis 2 was supported.
The first research question asked what are the most frequent strategy individuals with eating disorders use on-line to solicit social support. "Shared experiences" accounted for 28% of the messages generated by on-line support group users, followed by "requesting information"(13.9%), and "self-deprecating comments"(8%), (_2 (5, N= 490) = 444.70, p < .001).
The second research questions asked what is the most frequent theme addressed within on-line eating disorder discussion boards. Of the messages, about 20% of the messages fell in the "positive affect" category, followed by  "coping with weight" (14%), and "recalling symptoms" (13%), (_2 (9, N= 490) = 107.30, p < .001).



Discussion and Conclusions
According to the optimal match theory, eating disorders was argued to be controllable because the individual has control over the onset and termination of the stressor. Both hypotheses, derived from the theory, supported this claim.
Informational support was the most frequent type of social support provided on the discussion boards. This is consistent with the theory that suggest that individuals with eating disorders seek social support that focused on problem solving coping strategies (Cutrona & Russell, 1990). Providing advice and information regarding eating disorders creates a positive coping environment. However, emotional support was a close second type of social support provided.  It may be that some individuals seek more emotion-focused coping because they perceive themselves to have little control over their illness. Even if they do believe they have physical control over the illness, they would lack psychological confidence in their ability to do so. In addition, only 44% were not categorized as providing any support, this means that over half of the messages posted provided some kind of support to their on-line group members.
 Shared experiences were the most frequent social support soliciting strategy. This category involved phrases or statements of self-disclosure and described some type of experience, or questioned whether anyone shared a particular experience. Messages that were coded into this category usually sought out some type of informational support. This finding is consistent with the theory and the results of this study.
The primary theme addressed within the message postings was "positive affect", followed by "coping with weight". Positive affect was defined as providing another support group member with constructive feedback and encouragement. Support groups create a place where people are able to vent, share, and encourage others, indeed on-line support groups also facilitate this supportive environment. Another interesting finding, was that "reactions to the cultural pressures", which was defined as those message postings that involved the media's portrayal of beauty or other popular pressures to be thin was the least frequent category (1.2%). This finding reinforces the fact that eating disorders are a psychological disease and are not manifested from media effects.
Limitations
This study is not without some weaknesses. First, because of the difficulty of selecting a random sample of Internet support groups, these findings are only directly applicable to the Yahoo! site and may not represent the content of eating disorder support groups on other web sites. may be a product of those individuals that subscribe to the Yahoo site. In addition, because of the systematic sampling procedure, this study did not account for those individuals who posted multiple messages. Demographic data might be distorted by those individuals who posted a variety of messages during the year and may not be representative of the on-line population. Finally, it is unclear how the time of the messages was reported. Postings did not state which time zone was recorded.
Future Research
Several directions for future research emerge from these findings. First, future studies could be designed to provide additional support for the optimal match theory by assessing the other stress dimensions such as life domain or duration of consequence. Second, a quantitative study is needed to assess the effectiveness of these support networks over time.  This study suggests that, indeed, social support is being provided in over half of the messages posted on-line. However, it is unclear how successful these posting are. Also, an additional study designed to assess the attitudes, values, and beliefs of those individuals posting on-line messages. Measuring the attitudes towards these social supportive messages and the cathartic benefits to posting messages would give us a better idea of how individuals are using these support groups. Finally, future research should distinguish between those clinical diagnosed with an eating disorder and those individuals who are looking to lose a
 few pounds. Newly emerging "pro-anorexia" sites have attracted those individuals seeking "quick tips" to lose weight. However, future research should differentiate between these individuals and those who experience a mental disorder.
 References
Adelman, M. B., Parks, M. R., & Albrecht, T. L. (1987). Beyond close relationships: Support in weak ties. In T. L. Albrecht & M. B. Adelman (Eds.), Communicating social support (pp. 126-147). Newbury Park, CA: Sage.
Albrecht, T. L., & Adelman, M. B. (1987). Communicating social support. Newbury Park, CA: Sage.
Albrecht, T. L., & Adelman, M. B. (1984). Social support and life stress: New directions for communication research. Human Communication Research, 11, 3-32.
Albrecht, T. L. Burleson, B. R., & Goldsmith, D. (1994). Supportive communication. In M. L. Knapp & G. R. Miller (Eds.), Handbook of interpersonal communication (2nd ed., pp. 419-449). Newbury Park, CA: Sage
Albrecht, T. L., Burleson, B. R., & Sarason, I. G. (1992). Meaning and method in the study of communication and support: An introduction. Communication Research, 19, 149-153.
Buhrmester, D., Furman, W., Wittenberg, M.T., & Reiss, H. T. (1988). Five domains of interpersonal competence in peer relationships. Journal of Personality and Social Psychology, 55, 991-1008.
Burleson, B. R. (1994). Comforting messages. In B. R. Burleson, T. L. Albrecht, & I. G.  Sarason (Eds.), Communication of social support (pp. 3-28). Thousand Oaks, CA: Sage.
Burleson, B. R., Albrecht, T. L., Goldsmith, D. J., & Sarason, I. G. (1994). In B. R. Burleson, T. L. Albrecht, & I. G. Sarason (Eds.), Communication of social support (pp. xi-xxx). Thousand Oaks, CA: Sage.
Cobb, S. (1979). Social support and health through the life course. In M. W. Riley (Ed.), Aging from birth to death: Interdisciplinary perspectives (pp. 93-106). Boulder, CO: Westview Press.
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 310-357.
Cohen, S., Mermelstein, R., Kamarck, T., & Hoberman, H. M. (1985). Measuring the functional components of social support. In I. G. Sarason & B. R. Sarason (Eds.), Social support: Theory, research, and applications (pp. 73-94). The Hague, Netherlands: Martinus Nijhoff.
Crowther, J., Wolf, E. M., & Sherwood, N. E. (1992). Epidemiology of bulimia nervosa. In J. Crowther, D. L. Tennenbaum, S. E. Hobfoll, & M. A. P.  Stephens (Eds.), The etiology of bulimia nervosa: The individual and family context (pp. 1-26). Washington, DC: Hemisphere/Taylor & Francis.
Hays, R. B., Magee, R. H., & Chauncey, S. (1994). Identifying helpful and unhelpful behaviors of loved ones. AIDS Care, 6, 379-392.
Hill, C. A. (1987). Affiliation motivation: People who need people_ But in different  ways. Journal of Personality and Social Psychology, 52, 1008-1018.
Hsu, L. K. G. (1996). Epidemiology of the eating disorders. In J. Yager (Ed.), The Psychiatric Clinics of North America, 19, 681-700.
Kahn, R. L. (1979). Aging and social support. In M. W. Riley (Ed.), Aging from birth to death: Interdisciplinary perspectives (pp. 77-91). Boulder, CO: Westview Press.
Kessler, M., & Albee, G. W. (1975). Primary prevention. Annual Review of Psychology, 26, 7-15.
Klemm, P., Reppert, K., & Visich, L. (1998). A nontraditional cancer support group: The internet. Computers in Nursing, 16, 31-36.
Lin, N., & Peek, M. K. (1999). Social networks and mental health. In A. V. Horwitz & T. L. Scheid (Eds.), A handbook for the study of mental health: Social contexts, theories, and systems. New York: Cambridge University Press.
Magen, R. H., & Glajchen, M. (1999). Cancer support groups: Client outcome and the context of group process. Research on Social Work Practice, 9, 541-555.
Matheson, K., & Zanna, M. P. (1988). The impact of computer-mediated communication on self-awareness. Computers in Human Behavior, 4, 221-233.
North, C. L. (1997). Computer-mediated communication and social support among eating disordered individuals: An analysis of the alt.support.eating-diordered news group. Unpublished doctoral dissertation, University of Oklahoma.
Piran, N., Levine, M. P., & Steiner-Adair, C. (1999). Preventing eating disorders: A handbook of interventions and special challenges.  Philadelphia, PA: Brunner/Mazel.
Rice, R. (1984). Mediated group communication. In R. E. Rice Associates (Eds.), The New Media: Communication, Research, and Technology.  (pp. 129-154). Beverly Hills, CA:  Sage.
Rook, K. S. (1995). Support, companionship, and control in older adults' social networks: Implications for well-being. In J. F. Nussbaum & J. Coupland (Eds.), Handbook of communication and aging research (pp.437-463). Mahwah, NJ: Lawarence Erlbaum.
Rorty, M., Yager, J.,  Buckwalter, J. G., & Rossotto, E. (1999). Social support, social adjustment, and recovery status in bulimia nervosa.  International Journal of Eating Disorders, 26, 1-12.
Schaefer, C., Coyne, J. C., & Lazarus, R. S. (1981). The health-related functions of social support. Journal of Behavioral Medicine, 4, 381-406.
Sharf, B. F. (1997). Communicating breast cancer on-line: Support and empowerment on the internet. Women and Health, 26, 65-84.
Sparks, S. M. (1992). Exploring electronic support groups. American Journal of Nursing, 12, 62-65.
Turner, J. W., Grube, J. A., & Meyers, J. (2001). Developing an optimal match within online communities: An exploration of CMC support communities: An exploration of CMC support communities and traditional support.
Walther, J. B. (1994). Anticipated on-going interaction versus channel effects on relational communication in computer-mediated interaction. Human Communication Research, 20, 473-501.
Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23, 3-43.
Walther, J. B., & Burgoon, J. K. (1992). Relational communication in computer-mediated interaction. Human Communication Research, 19, 50-88.
Weber, K. D., & Patterson, B. R. (1996). Construction and validation of a communication based emotional support scale. Communication Research and Reports, 13, 68-76.
Weiss, R. S. (1974). The provisions of social relationships. In Z. Rubin (Ed.),  Doing unto others (pp. 17-26). Englewood Cliffs, NJ: Prentice-Hall.
Wiener, L. S. (1998). Telephone support groups for HIV-positive mothers whose children have died of aids. Social Work, 43, 279-286.
Wright, K. (2000). Computer-mediated social support, older adults, and coping. Journal of Communication, 25, 4-12.
Yalom, I. D. (1995). The theory and practice of group psychotherapy (4th ed.). New York: Basic Books.
Zelley, E. D. ( 2001). Desperately seeking support: What women with eating disorders want regarding social support from close friends. Paper presented at the National Communication Conference, Atlanta GA.
Zimmerman, S., & Applegate, J. L. (1994). Communicating social support in organizations. In B. R. Burleson, T. L. Albrecht, & I. G. Sarason (Eds.), Communication of social support (pp. 50-70). Thousand Oaks, CA: Sage.

 Footnote
1 There was no evidence that September 11th played a role in the results of this study.

 Table 1
Frequencies and Percentages of Social Supportive Strategies

Support                                 n               %
Informational                                   81              16.5
Emotional                                       76              15.5
Network                                 58              11.8
Tangible (Instrumental)                 32              6.5
Esteem                                          26              5.3
Not determined                          217             44.3
Total                                           490             100

_2 (5, N= 490) = 299.67, p < .001
 Table 2
Frequencies and Percentages of Soliciting Strategies

Strategy                                        n               %
Shared experiences                              137             28
Requests for information                        68              13.9
Self deprecating                                40              8.2
Statements of personal success          16              3.3
Statements of extreme behavior          3               .6
Not determined                          226             46.1
Total                                           490             100

_2 (5, N= 490) = 444.70, p < .001
 Table 3
Frequencies and Percentages of On-line Discussion Board Themes

Theme                                                   n               %
Positive affect                                         96              19.6
Coping with weight                                      71              14.5
Recalling symptoms of binging, purging and      65              13.3
starvation
Recommendations for prosocial behavior          48              9.8
Negative affect                                 41              8.4
Newcomer message                                        40              8.2
Coping with family, friends, co-workers         37              7.6
Unrelated to eating disorders                           36              7.3
Reactions to the cultural pressures                     6               1.2
Not determined                                  50              10.2
Total                                                   490             100

_2 (9, N= 490) = 107.30, p < .001
 Table 4
Operational definitions

Variable                                        Definitions/Examples
Theme
Coping with weight      A posting that discusses the challenges of gaining weight, losing weight, or eating. - "I gained three pounds last week, I had to use laxatives again"

Coping with external pressures
 from family, friends, coworkers        A posting that discusses the pressures from people other than themselves regarding their disorder. -"My mom was watching eat dinner, so I had to eat... then I threw up"

Reactions to the cultural pressures     A posting that discusses the media or
to be thin and standard of beauty       cultural pressures to be thin. -"I can't help
portrayed in the media  want to be thin after looking at all these models in the magazines"


Reminiscence of the symptoms    A posting that discusses the symptoms of an
of bingeing, purging, and starvation    eating disorder, including physical, psychological, or physiological. "I felt much better when I was binging because at least I was skinny and sad, and not fat and sad"

Negative affect A posting that expresses an overall negative attitude, including typing a sad face. - "I am so depressed I don't know if I can make it through the day, I had such a bad day, I don't think I can make it anymore"

 Table 4
Operational definitions

Variable                                        Definitions/Examples
Theme
Recommendations for prosocial
treatment       Including any posting that mentions prosocial behaviors, such as treatment facilities, posting messages, or attending a support group. - "The counselor I went to was awesome, you should try to go and not be afraid to be honest with these people, they can help"

Newcomer        A posting that includes a message from a first time user-"Hi, my name is Michelle, and this is the first time I have come to the support group" or a welcome message from a veteran user- "Welcome to the club, my name is Sara".

Positive affect A posting that encourages other users or provides constructive feedback. - "You are so strong, you are going to do great, good luck and let us know what happens"

Unrelated       A message posting that does not pertain to an eating disorder. "Hey, Kathy how are your kids? Didn't they start school today?"



Support Types
Informational Support   Phrases or statements suggesting advice, factual input, and feedback on specific actions Information about taking medication, going to a treatment facilities, or attending support groups,  "you need to go the doctor".

Table 4
Operational definitions

Variable                                        Definitions/Examples
Support types
Tangible (instrumental) phrases or statements suggesting specific goods or services, food, money, transportation, books. -"Carnation just came out with a new coffee creamer, hazelnut_it is fat free_try it!"


Esteem  Phrases or statements of respect for one's competency, abilities, skills, and intrinsic value "You have a wonderful ability to express yourself", "You have worked really hard, don't give up!", "you are a great person because you are able to see the best in other people" "you can make it through this"

Emotional       Phrases or statements of concern, caring, love, and empathy including compliments, "Katie, dear wonderful friend, I am so worried about you right now, you have no idea how much I worry and how sad I would be if something happened to you"

Network Phrases or statements indicating a sense of belonging or inclusion with people of similar interests and concerns, using words like "we" and "us" indicating a sense of community, including any mention towards the benefits of the support group system, "I just wanted to say that I am here, that there are others here who really care about you and want to be there for you. Please stay with us."

Table 4
Operational definitions

Variable                                        Definitions/Examples
Soliciting Strategies
Self-deprecating comments       Phrases or negative statements about oneself, minimizing (belittling) the individual or their problem, these posts indicate that the individual is not important or the post has many negative connotations that imply that the communication that occurred is not welcome or is undesirable  "I will shut up for awhile...thank you for letting me take up space I am a terrible person, I feel so fat, I even look fatter, a lot fatter and I'm not imagining it"

 Shared experiences     Phrases or statements of self-disclosure or describing some type of experience. These posts usually ask questions if anyone has shared a particular experience. They may include a question, but not necessary, "anyone out there know if bulimia can be a cause of irritable bowel syndrome?, "I was refused by all the insurance companies based on my history with bulimia_anyone else experience this?


Requests for information        Phrases or statements that request specific information without providing a great deal information about his or her own experiences, not concerned if someone can relate but rather looking for specific answers to specific questions, "I would like to hear more about mealtime in the ED program. Can someone pass along meal time advice given in treatment for these EDs?".



Table 4
Operational definitions

Variable                                        Definitions/Examples
Soliciting Strategies
Statements of personal success  Phrases or statements that reveal a personal success and positive improvements "Hi guys, yesterday I went to the therapist for the first time and she was cool! I woke up this morning and didn't want to be dead. I actually wanted to see if the sun was shining or not. Amazing. I weathered the storm".

Statements of extreme behavior  Phrase or statement describing something he or she has done or is planning to do that is above and beyond the "normal" range of daily activities for a person with an eating disorder, for example self-mutilation, suicidal, ideology, extreme restricting or exercising, excessive binging and purging, "I just swallowed many pills with a beer", " I would rather die than have this horrible feeling after eating. I guess I will have to OD on the laxative tonight". "I have such an urge to kill. To hurt something. To tear something apart. I could never inflict pain on anyone else, so I must do it to myself".